This is the submitted version of the article: “Holyoak, D.T., Holyoak, G.A., Lima, R.F. de, Panisi, M. and Sinclair, F. 2020. A checklist of the land () of the islands of Sao Tome and Principe, with new records and descriptions of new taxa. Iberus, 38 (2): 219-319.”

This version has not been peer-reviewed and is only being shared to comply with funder requirements.

Please do not use it in any form and contact the authors (e.g.: [email protected]) to get access to the accepted version of the article.

1

New species and genera and new island records of land snails (Gastropoda) from the islands of São Tomé and Príncipe

Nueva especies de ....

David T. HOLYOAK1, Geraldine A. HOLYOAK1, Ricardo F. de LIMA2,3, Martina PANISI2 and Frazer SINCLAIR3,4

Recibidio el ...

ABSTRACT

Seven species of terrestrial Gastropoda are newly described from the island of São Tomé and six more from the island of Príncipe. The genera involved are Chondrocyclus (Cyclophoridae), Maizania and Thomeomaizania (Maizaniidae), Pseudoveronicella (), Nothapalus (: subfamily undet.), Gulella and Streptostele (), (), Afroconulus (), Principicochlea gen. nov., Principotrochoidea gen. nov., Thomithapsia gen. nov. and Thomitrochoidea gen. nov. (). Most of these are from natural forest and are likely to be single- island endemics. Apothapsia gen. nov. () is also described to accommodate two previously known species. Additional new island records are of ten species on São Tomé, one on Príncipe alone and two more on both islands. These include six species of "microgastropods" with wider ranges in tropical Africa that are likely to be hitherto overlooked parts of the indigenous fauna and six anthropogenic introductions; Pseudopeas crossei previously known only from Príncipe and Bioko is newly recorded on São Tomé. More detailed taxonomic notes are provided on identification and delimitation of species and genera in Cyclophoridae/ Maizaniidae, Veronicellidae, Achatinidae (for Petriola), Helicarionidae, and on the forgotten but apparently valid Pseudoveronicella thomensis (Veronicellidae). The first comprehensive checklists since the works of Germain in 1916 are presented for the terrestrial Mollusca of both islands, comprising 86 species in total, with 59 known for São Tomé and 45 for Príncipe.

RESUMO

Sete novas espécies de gastrópodes terrestres são descritas da ilha de São Tomé e seis da ilha do Príncipe. Os géneros envolvidos são Chondrocyclus (Cyclophoridae), Maizania e Thomeomaizania (Maizaniidae), Pseudoveronicella (Veronicellidae), Nothapalus (Achatinidae: subfamília indet.), Gulella and Streptostele (Streptaxidae), Truncatellina (Truncatellinidae), Afroconulus (Euconulidae), Principicochlea gen. nov., Principotrochoidea gen. nov., Thomithapsia gen. nov. e Thomitrochoidea gen. nov. (Urocyclidae). A maioria destes registos foram feitos em habitats de floresta natural e provavelmente são endémicos de uma única ilha. Apothapsia gen. nov. (Helicarionidae) também é descrito para acomodar duas espécies já conhecidas. 10 espécies adicionais são novos registos para São Tomé, uma para o Príncipe e duas para ambas as ilhas. Estes incluem seis espécies de "microgastrópodes" com ampla distribuição na África tropical, que provavelmente são parte da fauna indígena que tinha passado despercebida, e seis introduções antropogénicas; Pseudopeas crossei anteriormente conhecida apenas do Príncipe e Bioko é um novo registo para São Tomé. São fornecidas notas taxonómicas mais detalhadas sobre a identificação e delimitação de espécies e géneros de Cyclophoridae/ Maizaniidae, Veronicellidae, Achatinidae (para o género Petriola), Helicarionidae e para a

2

esquecida, mas aparentemente válida Pseudoveronicella thomensis. São apresentadas as primeiras listas compreensivas de moluscos terrestres para ambas as ilhas desde os trabalhos de Germain em 1916, compreendendo 86 espécies no total, com 59 conhecidas para São Tomé e 45 para o Príncipe.

RESUMEN

Se describen x especies nuevas de Gastropoda terrestre de la isla de Santo Tomé (to be completed)

1Quinta da Cachopa, Rua da Barcoila no. 274, 6100-014 Cabeçudo, Portugal. Corresponding authors, e-mail: [email protected] 2Centre for Ecology, Evolution and Environmental Changes (Ce3C), Faculdade de Ciências, Universidade de Lisboa, 1749-016 Lisboa, Portugal. 3Departamento de Biologia , Faculdade de Ciências, Universidade de Lisboa, 1749- 016 Lisboa, Portugal. 4Fauna & Flora International, Cambridge, U.K. 5Fundação Príncipe, Santo António, Príncipe Island, Republic of São Tomé and Príncipe. ______

3

INTRODUCTION

The Republic of São Tomé and Príncipe comprises two mountainous islands lying close to the Equator in the Gulf of Guinea, off the west coast of central Africa. Both islands form part of the Line of volcanos, and both arose due to volcanic activity during the Tertiary. They rise from oceanic depths and have always been isolated from the nearest African coasts, which are no closer than 220 km and 255 km distant at the present day. Their natural vegetation was tropical , although cultivation has replaced most of it on the lower ground since human colonisation began in the fifteenth century (JONES & TYE, 2006). Both islands support rich land-snail faunas with a high proportion of endemic species. GASCOIGNE (1994a, 1994b) reported the faunas as comprising 39 species for São Tomé and 32 species for Príncipe, although he never published a full list for either island. Among these, he calculated for São Tomé that 25 species (64.1%) were single island endemics, with 19 (59.4%) for Príncipe; an additional 5 and 6 species respectively for each island were regarded as endemic to the Gulf of Guinea islands as a whole (i.e. present on more than one of the islands of São Tomé, Príncipe, Bioko or Annobon). GASCOIGNE (1993, 1996) also published useful bibliographies of the fauna of the islands. Most of the land-snail species were discovered and named during the nineteenth century, in publications by RANG (1831), MORELET (1848, 1858, 1868), DOHRN (1866a, 1866b), GREEFF (1882), NOBRE (1886, 1891, 1894) and GIRARD (1893a, 1893b), with reviews by CROSSE (1868, 1888a, 1888b) and Girard. In the early twentieth century, GERMAIN (1908, 1912a, 1912b, 1915, 1916) reported on additional large collections of specimens, but the proportion of new species he found was lower, so one might gain the impression that faunal exploration had almost been completed. After this there was a long break in activity, until GASCOIGNE (1994a) became resident on São Tomé and reported new discoveries of a few species that had been introduced to the islands. However, the paper by GERMAIN (1916) was the last to give a comprehensive critical checklist of the land Mollusca of either of the islands. DTH and GAH visited both islands in December 2013 and made small collections of land Mollusca, and subsequently prepared a review of the endemic Rhysotina of São Tomé (HOLYOAK & HOLYOAK, 2016). RFL and MP made several longer visits to the islands over the next few years, partly to study Archachatina, during which small numbers of other molluscs were obtained. Unidentified taxa from the collections made in 2013 appeared to represent undescribed species, so DTH and GAH returned to both islands in November- December 2018 in order to collect more systematically. Their fieldwork on Príncipe was carried out with assistance from FS and his co-workers from the Fundacão Príncipe and these continued to collect molluscs in other areas during 2019. RFL and MP also continued to work on both islands in 2018 and 2019, resulting in more specimens found during studies focussed on Archachatina. This paper describes the 13 new species discovered during all of this fieldwork, along with other material representing new distributional records for either island. We found that very small snails (<3 mm) had often been overlooked by earlier workers, doubtless because they mainly occur at very low densities, sieving techniques were not used in the past and the weak daylight reaching the ground inside the makes direct searching unproductive. However, some taxa with larger shells had also been overlooked, possibly because they resembled known species. Research on the material collected is continuing in order to clarify and update the of many of the endemic species from anatomical and other characters, a process made more difficult by the poor level of taxonomic understanding of the faunas of neighbouring continental countries. Because nothing better

4

exists to fill the need for a full checklist, we give provisional full species lists for both islands in Table 1.

MATERIAL AND METHODS

Fieldwork was carried out by DTH and GAH on São Tomé from 1-7 and 11-13 Dec. 2013 and 24 Nov.-2 Dec. and 8-12 Dec. 2018; on Príncipe on 9-10 Dec. 2013 and 3-6 Dec. 2018. In 2018 we were assisted in the field on some days by RFL, MP and FS, along with local guides and other helpers, and these authors also made small collections independently. Fieldwork sites were selected by DTH and GAH to represent a range of molluscan habitats, but with a preference for native forest in good condition. At each site we attempted to find and record a full list of all species of non-marine Mollusca present, including , and to always collect samples of those that could not be reliably identified in the field. In wooded habitats the searching generally included foliage of herbs, ferns and saplings (extending up to five metres above the ground by bending thin saplings over), leaf-litter, trunks and exposed roots of trees, and beneath dead branches or any rocks lying on the ground. Locations were recorded with a hand-held GPS (Garmin eTrex high sensitivity) that was accurate to within 5 m horizontally and vertically on open ground. This GPS remained useable inside tall forest, but its accuracy was considerably reduced, especially in measuring altitude. Coordinates for localities have been converted to decimal degrees. Paired sieves (meshes 2.0 and 0.5 mm) were regularly used in the field, although many of the sites sampled inside forests were so dark that the fine sieved fraction ("fines") was usually taken away in labelled plastic bags for drying and later searching. Really wet ground litter was also sometimes removed in bags for later drying and sieving. Since small snails were found to be generally scarce when present at all in the forest leaf-litter, sieving was soon focussed on the few places where accumulations of larger shells provided calcium and apparently increased the concentrations of small snails. On São Tomé the anvil sites of the São Tomé Thrush Turdus olivaceofuscus provided these, where many tens of broken shells of Rhysotina hepatizon (Gould, 1845) were often accompanied by smaller numbers from R. welwitschi (Morelet, 1866), Petriola spp. and occasional immatures of Archachatina bicarinata (Bruguière, 1792). In forest on Príncipe, shells concentrated by Blue-breasted Kingfishers Halcyon malimbica dryas provided similar but less productive sieving opportunities, as did heaps of shells of Archachatina marginata collected for human food. The kingfisher "anvils" there included piles of old shells of Columna columna (O.F. Müller, 1774); "these are broken against rocks where hundreds of smashed shells may pile" (LEVENTIS & OLMOS, 2009: 86). Living molluscs were drowned in water overnight then stored in ims or 80% ethanol, with bodies of the larger taxa being pulled wholly or partly from the shells. Most empty/dead shells collected into specimen tubes were also placed in ims or 80% ethanol, then cleaned and dried later, although some of the largest were dried during the fieldwork. Samples of "fines" dried during fieldwork were mostly searched later (in Portugal) using a low-power stereo-microscope with large field of view and good lighting. Descriptions of shells were made using Meiji RZ Series stereo-microscopes with fibre- optic light sources. Measurements were made with an eyepiece micrometer, drawings with assistance from a Meiji drawing tube, microphotography using Infinity (Lumenera Corporation) equipment. Shell whorls were counted following the method shown by KERNEY & CAMERON (1979: 13). Dissections were made under ims using the same equipment, with genitalia drawn after they were removed from bodies. Proximal and distal refer to positions relative to the gonad.

5

Abbreviations: alt.: altitude above sea level; AB: aperture breadth; AH: aperture height; B: shell breadth; bod: bodies in ims; CGAH: Collection of G.A. and D.T. Holyoak; DTH: D.T. Holyoak; FS: F. Sinclair & co-workers; GAH: G.A. Holyoak; H: shell height; ims: Industrial Methylated Spirit (80%); leg.: collected by; MP: M. Panisi; NHMUK: The Natural History Museum, London, U.K.; n.v.: original not verified; RFL: R.F. de Lima; sh: shells; SMF: Senckenburg Museum, Frankfurt, Germany; spm: specimens in ims; syn.: synonym; TL: type locality; UB: umbilicus breadth. Additional abbreviations used for anatomy and external parts of body in the figures are as follows ‒ aa: atrial appendage; ag: albumen gland; am: muscle attached to genital atrium; at: annular thickening on penis (in Pseudoveronicella); bc: bursa copulatrix; bcd: duct of bursa copulatrix; ch: caudal horn; chd: common hermaphrodite duct; cm: part of columellar muscle; cs: cylindrical sac (beside chd in Streptostele); dep: distal part of epiphallus; dg: digestive gland; em: embryo developing in oviduct; ep: epiphallus; epc: caecum on epiphallus; f: caudal fossa; fo: free oviduct; ga: genital atrium; go: gonad; gp: external genital pore; gr: deep groove beneath head (Helicarionoidea); lc: lateral caecum inside os; lt: lower tentacle; L1, L2, etc.: mantle laps (Helicarionoidea); m: mouth; ma: dorsal surface of mantle inside shell; ms: muscle attaching bc to pr; os: thin outer sheath of penis; osp: outer sheath of penis complex (in Pseudoveronicella); p: penis; pc: penial caecum; pcg: pericaudal groove; pep: proximal part of epiphallus; pi: pilaster; ps: penis sheath; pg: penial gland; pr: prostatic part of sod; prm: penis retractor muscle; rmaa: retractor muscle on atrial appendage; ro: right ommatophore; saa: thin sheath of atrial appendage; sod: spermoviduct; st: stimulator arising in atrial appendage; th: thickening on vas deferens; ut: upper tentacle; v: verge; va: vagina; vd: vas deferens. All specimens collected by DTH and GAH from 2013-2018 except holotypes are currently in CGAH.

TAXONOMIC PART & RESULTS

The supra-generic classification used here mainly follows BOUCHET ET AL. (2017).

CYCLOPHORIDAE J.E. Gray, 1847 or MAIZANIIDAE Tielecke, 1940

Two endemic species from São Tomé named and described in Cyclophorus by NOBRE (1866) were placed in separate genera as Afroditropis molleri and Maizania (Thomeomaizania) vandelli by BEQUAERT & CLENCH (1936). TIELECKE (1940) used anatomical characters of other species to recognise two separate families, the Cyclophoridae and Maizaniidae. VAN BRUGGEN (1986) followed Tielecke's recognition of two families and treated Thomeomaizania as a full genus within the Maizaniidae, despite the absence of anatomical information for either of the two species hitherto reported from São Tomé. VAN BRUGGEN (1986: 359) attempted to define characters of the shell and operculum to discriminate Maizaniidae from Cyclophoridae at family level, but it appears that he was only partly successful in doing so. He noted (following BOSS, 1982: 979) "that the operculum of the Maizaniidae is always thin, corneous and simple", while that of Cyclophoridae is "corneous or calcareous, generally multispiral, and occasionally with external processes". However, the two species already known from São Tomé and two new species named below all have a similar operculum: it is concave externally, with a small smooth central circle (nucleus) surrounded by a lamellar ridge which spirals tightly for up to 11 full turns. The edge of the operculum has a rounded concave groove in its outer half, a flat edge on its inner part. The whole structure usually appears corneous and typically whitish or buff or pale brown in colour; in Thomeomaizania vandellii it is whitish externally in some populations

6

but brown in other samples of the same species, possibly reflecting varying extents of calcification. Thus it matches the Cyclophoridae type rather than the Maizaniidae type, despite VAN BRUGGEN (1986: 359-363) having placed this genus in Maizaniidae. However, the genus Thomeomaizania Bequaert & Clench, 1936 appears to have been defined by VAN BRUGGEN (1986: 364) on the basis of a single shell that has lost most of its periostracal sculpture (see below), so its affinities are likely to have been misinterpreted. Shell characters of the four species now known from São Tomé seem to span much of the diversity known hitherto in both families, with respectively, predominantly spiral sculpture of three strong ridges, predominantly radial sculpture plus either two strong or at least five weak spiral keels, or a smooth surface without obvious spiral or radial structures. Hence, it seems likely that a taxonomic revision using molecular-genetic characters alongside those of genital anatomy and shells will be needed in future to produce a sound classification at generic and family levels. As a provisional arrangement in the absence of both molecular and anatomical data, we therefore name the two new species from São Tomé and a third from Príncipe in the existing genera to which their shells show the strongest resemblance. Nevertheless, the similarity in the opercula may imply that at least the four species from São Tomé represent a single endemic radiation of forms; if so, they should probably all be placed in Thomeomaizania. A key to identify shells of the five species now known from the islands is presented at the end of this section.

Chondrocyclus Ancey, 1898

Chondrocyclus inexspectatus sp. nov. G. Holyoak & D. Holyoak (Fig. 1A)

Type material: Named and described only from the unique holotype shell, NHMUK reg.no., collected 4 Dec. 2018 by GAH at site 18-12. Type locality: Príncipe Island, path to Santa Joaquina, N1.6157° E7.3976°, 216 m alt., tall native forest with understorey of saplings, found on leaf-litter at base of steep flushed bank. Etymology: The species epiphet inexspectatus (Latin, meaning unexpected) refers to our surprise at finding a distinct undescribed cyclophorid at an easily accessible locality on an island where land snails were well studied historically, and further surprise on later realising that it may represent a different genus to the Cyclophoridae known from São Tomé.

Description: Based on holotype shell which is apparently mature and in fresh condition: H 2.57 mm, B 2.82 mm, H/B 0.91; AH 1.25 mm, AB 1.09 mm, AH/AB 1.14; UB 0.72 mm (UB/B 25.5%); whorls 3.6. Shell depressed-conical, with rapidly expanding whorls descending regularly. Whorl profile rounded, with deep suture. Aperture nearly round (slightly higher than broad). Peristome thin, not reflected, curving smoothly in ventral view; continuous where it meets penultimate whorl and narrowly fused with latter externally; in oblique view, a shallow recess apparent around umbilicus. Umbilicus comparatively large and open, symmetrical, exposing interior of whorls of spire. Protoconch raised, of ca 1.5 whorls, with microsculpture of minute low rounded papillae. Teleoconch of much of upper part of spire partly concealed (encrusted), but with six spiral cords projecting, the three cords on outer part of penultimate whorl bearing well-spaced perpendicular hairs resembling those on body-whorl. Upper and outer surfaces of body-whorl with ca nine spiral cords, ± evenly spaced, with raised crests, reaching peristome edge, the four outermost bearing rather evenly- spaced but sparse hairs (hairs 0.11-0.18 mm long, brownish, mainly straight, very narrowly conical and tapering to tip, mostly perpendicular to shell surface, but those near shell aperture recurved from base); finer radial sculpture of closely-spaced riblets is continuous across the cords and the wider spaces between them, these riblets having much weaker spiral

7

lines just visible at ×56 magnification. Underside of body-whorl with five spiral cords, plus four more in umbilicus, all ± evenly spaced, extending outwards to edge of peristome, the external cords with lamella-like crests and the two outermost with a few hairs as on upper part of body-whorl; whole undersurface also with finer radial sculpture, resembling that on uppersurface of whorl. Protoconch whitish; teleoconch blackish-brown, glossy; interior of aperture dark brown with external cords showing as pale lines, the thin shell slightly translucent. Operculum closing interior of body-whorl 0.2 whorl back from aperture (perhaps attached to dried body internally), with similar bistratose structure to that noted in family account above revealed by damage along one edge; externally whitish, concave, darker smooth central ‘nucleus’ forming ⅓ of diameter; raised lamella spiralling outwards from nucleus with six closely appressed near-circular coils visible, but true number probably greater because surface apparently damaged or eroded. The exterior of the body, genital anatomy, radula, etc., remain undescribed. Discussion: The identity of this single small prosobranch specimen as a distinctive taxon new for Príncipe was apparent in the field. However, close searching around the immediate area where it was found failed to reveal any more individuals and two large bags of leaf-litter taken away, dried and then sieved were also unproductive. The present species is placed in the genus Chondrocyclus on the basis of presence on the body-whorl of strong spiral keels, reticulate sculpture and a hirsute periostracum (VAN BRUGGEN, 1986: 360, 374-377; HERBERT & KILBURN, 2004: 90-92). However, what remains of the operculum apparently differs markedly from the description of its external appearance in South African species of this genus given by HERBERT & KILBURN (2004: 90) of a "spirally coiled fringe of strong, flattened, usually fused bristles; bristles project around the edge, forming a tight seal …" Furthermore, the periostracum is minutely hirsute on immatures of Thomeomaizania gascoignei (see below).

Maizania Bourguignat, 1889

Maizania furadana sp. nov. G. Holyoak & D. Holyoak (Fig. 1D)

Type material: Holotype, shell (H 2.59, B 1.77 mm) in NHMUK reg.no., collected 26 Nov. 2018 by GAH & DTH at site 18-5. Type locality: São Tomé Island, near Ponta Furada, N0.23665° E6.46775°, 240 m alt., on flushed, steep rock of trackside cutting, part-shaded by secondary forest. Paratypes: 30 shells from the type locality, mainly with opercula. Etymology: The species epiphet furadana is an adjective constructed from the name of Ponta Furada, a coastal headland near the type-locality.

Description: Apparently mature shells: H 1.83-2.59 mm, B 1.44-1.95 mm, H/B 1.15- 1.46; AH 0.72-1.01 mm, AB 0.72-0.99 mm, AH/AB 0.96-1.02; UB 0.23-0.51 mm (UB/B 15.7-26.2%); whorls 3.5-4.3. Shell conical with whorls rapidly expanding and descending; whorls evenly rounded in profile, separated by very deep suture; body-whorl descending gradually to aperture. Umbilicus narrow, interior of top of spire just visible, in largest shells becoming excentric at body whorl. Umbilicus with 0-3 distinct low narrow ridges formed of periostracum, with rough tops, well-separated from each other, all emerging to almost reach the outer edge of the peristome. Aperture almost circular (varying from slightly higher than wide to slightly wider than high, sometimes a slight blunt point at top, near where exterior touches penultimate whorl; peristome thin, not reflected, in oblique basal view with shallow recess adjacent to umbilicus. Protoconch minutely rough with low rounded papillae; teleoconch appearing almost smooth at low magnification but with weak, close, rather

8

irregular radial growth lines; at high magnification (×56) faint close spiral lines are just visible. Protoconch of 1.3-1.5 whorls whitish, teleoconch externally ± dark brown, sometimes nearly blackish, glossy; interior of aperture on mature shells whitish and almost opaque, although shell rather thin. Operculum closing interior of body-whorl ca 0.1-0.3 of whorl back from aperture, attached internally to foot of dried body; general structure as described above under family heading; concave externally, whitish to light yellow-brown; smooth ‘nucleus’ forming central ¼-⅓ of width, remainder with narrow raised lamella or ridge spiralling outwards in ca 11 tightly appressed near-circular coils. External coloration of the body of living was mainly pale grey. Their tentacles were long, slender and blackish, remaining visible through the translucent shell. Eye spot at base of and external to each tentacle was black. The operculum carried on top of the tail appeared whitish. Moist shells containing bodies of the living snails appeared dark olive- brown to blackish. Characters of the genital anatomy remain undescribed. Dried bodies are present in numerous paratype shells, so study of the radula should be possible in future. Discussion: The new species is known only from its type-locality. The living snails were resting or moving on moist dead leaves and stems in places with dripping water on the volcanic rock of the old cutting. This species is provisionally placed in the genus Maizania on the basis of the shell lacking distinct sculpture. Following VAN BRUGGEN (1986: 360-362), its small size (<4 mm) would also imply it belongs in subgenus Micromaizania Verdcourt, 1964. Again, following VAN BRUGGEN (1986: 359), the present species would be assigned to Maizaniidae rather than Cyclophoridae on the basis of the absence of "Major spiral sculpture in the form of keels, ridges or cords". Nevertheless, the operculum in not thin, corneous and simple which he regarded as characteristic for that family. Evidence from elsewhere also shows that lack of spiral shell sculpture is not always a reliable character pointing to Maizaniidae, since HERBERT & KILBURN (2004: 92) characterised Chondrocyclus exsertus (Melvill & Ponsonby, 1903) of the Cyclophoridae from KwaZulu-Natal as having a smooth shell but a duplex operculum.

Thomeomaizania Bequart & Clench, 1936

Thomeomaizania gascoignei sp. nov. G. Holyoak & D. Holyoak (Fig. 1B)

Type material: Holotype, shell (H 3.25, B 2.84 mm) in NHMUK reg.no., collected 4 Dec. 2013 by GAH & DTH at site ST4. Type locality: São Tomé Island, path to Lagoa Amélia from Bom Sucesso Botanical Garden, N0.2786° E6.6000°, ca 1300 m alt., in montane forest on slope with understorey of saplings, 23 sh (mainly live-collected, along with 36 sh of T. vandellii). Paratypes: 22 sh from type-locality collected with holotype on 4 Dec. 2013, GAH & DTH at site ST4. Also: 7 Dec. 2013, path to Lagoa Amélia from Bom Sucesso Botanical Garden, N0.2825° E6.5969°, ca 1400 m alt., montane forest on ridge, with closed canopy & understorey of saplings, GAH & DTH ST8, 1 sh (live-collected, along with 11 sh of T. vandellii); 9 Dec. 2018, ditto, ca N0.2823° E6.5964°, 1290-1415 m alt., montane forest, GAH & DTH 18-21, 19 sh (live-collected, along with 55 sh of T. vandellii); 10 Dec. 2018, near Lagoa Amélia, N0.2812° E6.5908°, 1416 m alt., low moist montane forest around infilled crater-lake, DTH, MP & FS 18-25, 3 sh (live-collected, along with 4 sh of T. vandellii). Etymology: The species epiphet honours the late Angus Gascoigne, a former resident on São Tomé whose publications helped to revive interest in the land Mollusca of the islands after the decades of inactivity throughout the mid-twentieth century.

9

Description: Apparently mature shells: H 2.65-3.31 mm, B 2.18-2.84 mm, H/B 1.14- 1.34; AH 1.05-1.32 mm, AB 1.09-1.32, AH/AB 0.96-1.00; UB 0.43-0.62 mm (UB/B 19.2- 23.0%); whorls 3.5-4.6. Shell conical, whorls increasing and descending rapidly, sutures deep. Whorl profile externally with several weak spiral keels: three at and above periphery of body-whorl, two more beneath periphery, all separated by flatter bands of shell and ending just before the peristome edge. Aperture almost round, with peristome thin to very thin, plane, not reflected, sometimes (in subadults?) interrupted at point of contact with penultimate whorl, in oblique view from below showing shallow recess adjoining umbilicus. Umbilicus narrow, internally just exposing whorls of spire, becoming somewhat eccentric at body-whorl in larger shells. Exterior of lowest part of body-whorl in umbilicus with up to five more weak spiral keels, each forming a thin ridge of periostracum extending outwards to end close to edge of peristome lip, the crest of each ridge with tiny elongate pointed scale (up to 0.07 mm long) arising where ridge meets radial sculpture. Protoconch (up to whorl 1.5) with closely spaced low weak radial ridges and minute low rounded papillae. Teleoconch whorls 2-3 with ± regular fine radial ribs; an immature shell has minute (0.06 mm) hair-like projections on the periostracum in a row where the ribs approach the suture. Lower whorls of spire and body-whorl with microsculpture of minute closely spaced radial riblets crossed by closely-spaced minute (×56) spiral lines; this sculpture interrupted at intervals by prominent radial bands formed of 3-5 much stronger radial ribs grouped together, each of them bearing erect scales where it intersects a spiral keel (although scales fragile and soon lost). Underside of body-whorl lacking strong radial ribs, but with closely spaced pattern of minute radial riblets and even finer spiral lines (just visible at ×56). Protoconch white, teleoconch brown to blackish-brown, glossy on areas of upperside lacking strong ribs and on most of underside. Interior of aperture brown, the shell rather thin and somewhat translucent. Operculum inside aperture 0.2-0.3 whorl in from peristome, attached internally to foot of dead body. See family heading for general characters of opercula, which externally whitish to yellow-brown, concave, with flat central ‘nucleus’ forming less than ¼ of total diameter, surrounded by a sharp lamella spiralling outwards in 10-11 closely appressed near-circular coils. No information was recorded on characters of the exterior of the body or on genital anatomy because the specimens were intermixed with those of T. vandellii and not recognised as distinct until after fieldwork was completed. Discussion: Our shell samples clearly demonstrate that two superficially similar species occur together in montane forest on São Tomé which both resemble Thomeomaizania vandellii (Nobre, 1886). However, we found initially that there was scope for serious doubts about which of the two species should now be identified as T. vandellii, but believe these have subsequently been resolved as discussed below. Hence, the second species is named here as Thomeomaizania gascoignei sp. nov. and the Key below sets out the characters used to separate it from T. vandellii. VAN BRUGGEN (1986: 364-366, figs 4-6) selected a lectotype for Cyclophorus vandellii Nobre, 1886 as BM [now NHMUK] 98.1.22.11, "Presd. by Mr. Aug. Nobre", and redescribed and figured this specimen. CROSSE (1888a: 27-28, pl. 1 figs 2, 2a) also described and figured "Cyclophorus Vandellii, Nobre" independently of the original valid description by NOBRE (1886: 14-15). Besides the specimen sent to the BM and another to Crosse, Nobre is likely to have retained others in Portugal, but his collection went to the Lisbon Museum (DANCE, 1986: 220) where it was apparently lost in the fire which destroyed the Zoology Collection in March 1978. Comparison of van Bruggen's detailed re-description and good figures of the lectotype with the descriptions by Nobre and Crosse (and the good figures in CROSSE, 1888a) suggested at first sight that the lectotype might represent a different species: it has much

10

weaker vertical ribbing, less prominent spiral keels, and no sign of projections ("tubercles") arising from the spiral keels (the last difference being noted by VAN BRUGGEN (1986: 366) himself), and these features appear more similar to the characters of T. gascoignei. The description by Nobre suggested more prominent ribs and keels than did van Bruggen's new description ("coberta de uma epiderma lamellosa e sobreposta em camadas espaçadas verticaes; and [for body whorl profile] "ultima fortemente subquadrangular"); Crosse's description likewise emphasised strong keels and prominent sculpture ("dernier tour muni de 3 rangées de tubercules saillants, dont la plus rapprochée de la base est la plus faible, et qui le font paraitre comme bicaréné") and his figures showed these characters well (cf. our Fig. 1B and C) . Nonetheless, VAN BRUGGEN (1986) referred to periostracum wrinkles ... "being crowded and forming a fringe around the umbilicus" (p. 365, repeated on p. 366) and these apppear to be indicated in his figs. 4 and 6. This "fringe" is a decisive character, formed of a comb-like or lamella-like band curving around the edge of the widest part of the umbilicus on the underside of the body whorl, formed of elongate closely spaced periostracal projections (see our Fig. 1C, middle photo). T. gascoignei lacks this, having instead two to five much lower spiral ridges deeper within the umbilicus (Fig. 1B middle and lower photos). Hence it appears that the lectotype does indeed represent C. vandellii, but with a specimen having most of the periostracal sculpture worn, corroded or rubbed off the body whorl and spire. We can match this shell with a proportion of our recently collected specimens. As noted above, T. vandellii was found living with T. gascoignei at all four sites where the latter was recorded, at 1300-1416 m altitude in montane forest. However, T. vandellii has a much wider range on São Tomé, extending from 1416 m alt. beside Lagoa Amélia to the southern end of the island at low altitudes (e.g. 11 Dec. 2013, by EN2 ca 2 km S. of Monte Mário "praia", N0.0694° E6.5522, ca 101 m alt., roadside bank at foot of slope with secondary forest, GAH & DTH ST11, 15 sh). Maizania (Thomeomaizania) was named by Bequaert & Clench, 1936, Rev. Zool. Bot. Afr., 29, p. 99, with type species by original designation and monotypy Cyclophorus vandellii Nobre, 1866. VAN BRUGGEN (1986: 364) treated Thomeomaizania as a full genus within the Maizaniidae, which he evidently characterised on the basis of the worn lectotype shell discussed above, showing: "small size and turbinate shape (l/d about 1.00), the presence of weak costulae and an intensely wrinkled, deciduous periostracum, which shows two faint keels." This seriously misrepresented the strong ribbing and two prominent spiral keels of an unworn shell of the type species, which, following the discussion and key by VAN BRUGGEN (1986: 359-360), would instead lead to its identification as a genus in the Cyclophoridae.

Key to species of Cyclophoridae and Maizaniidae of São Tomé and Príncipe

For use with unworn adult shells, if necessary after carefully cleaning away adhering soil.

1. Shell wider than high (Fig. 1A); umbilicus ca 25% of shell breadth; periostracum with longer hair-like projections (>0.1 mm long); Príncipe Island only Chondrocyclus inexspectatus sp. nov. -. Shell usually higher than wide; umbilicus <25% of shell breadth; periostracum lacking hair-like projections, or with these restricted to minute hairs (<0.1 mm long) on immature shells; São Tomé Island only 2.

2. Upper surface of shell with sharp radial (vertical) ribs bearing scales (that may form rough papillae at crests of the spiral keels) (Fig. 1 B, C) 3.

11

-. Upper surface of shell without sharp radial (vertical) ribs bearing scales, ± smooth (Fig. 1D) 4.

3. Body-whorl rounded with 5 weak spiral ridges (3 at and above periphery, 2 below it) (Fig. 1B); 2-5 ridges prominent within umbilicus (behind columellar side of aperture); comb-like lamella lacking from entrance to umbilicus; most shells not heavily encrusted with soil Thomeomaizania gascoignei sp.nov. -. Body-whorl sub-quadrangular (with 2 strong spiral keels above periphery); underside of shell and interior of umbilicus lacking spiral ridges (Fig. 1C); ridges lacking within umbilicus but entrance to umbilicus fringed externally by curved comb-like lamella formed of narrow scales; shells often encrusted with soil Thomeomaizania vandellii

4. Shell with three spiral cords on upper surface Afroditropis molleri -. Shell lacking spiral cords on upper surface (Fig. 1D) Maizania furadana sp. nov.

VERONICELLIDAE J.E. Gray, 1840

The review of African Veronicellidae by FORCART (1953) recorded only Pseudoveronicella liberiana (Gould, 1850) for São Tomé and Príncipe. Our study recognises four species of the family from the islands, adding a species which is newly described here, a poorly-known taxon rescued from synonymy and an alien that may be a relatively recent arrival. A key to identify these is given below. We mainly follow the terminology of FORCART (1953: 16-18) for the external morphology and the genital organs of this family (with upperside of body termed notum, underside comprising sole of the foot with hyponotum forming band on each side of it, separated from foot by a pedal groove). The methodology he suggested (p. 19) for dissection was also adopted, consisting of opening the skin by cutting along the left-hand pedal groove, before turning back the skin over the full length of the body. Male genitalia were located along the right-hand side of the buccal mass and just proximal to it, with the oesophagus often passing through the digitiform branches of the penial gland. The male genitalia were removed for study, the penial sheath then being opened longitudinally from the base upwards to expose the verge. Additional anatomical terminology translated from GOMES (2007: 4) is adopted for the parts of the vas deferens: anterior vas deferens ("ducto deferente anterior") being the section extending from near the external male genital pore to the proximal end of the penis, middle vas deferens ("deferente médio") the part passing concealed within the body wall from near the external male pore proximally to near the posterior end of the body.

Laevicaulis Simroth, 1919

Laevicaulis alte (A. Férussac, 1821)

Vaginulus alte A. Férussac, 1821, Tableaux systématiques des animaux mollusques ..., part II, Limaces, p. 14. TL "environs de Pondichéry" [= Puducherry, India]. The publication date (based on KENNARD, 1942) of p. 14 in the quarto edition which gave the original description was 6 April 1821 in Livraison 9, not 1822 as given on MolluscaBase (http://www.marinespecies.org/aphia.php?p=taxdetails&id=1058339), accessed 5 Sept. 2019. New to Príncipe: 3 Dec. 2018, just S. of Santo António on road to Bela Vista, N1.6307º E7.4179º, ca 24 m alt., edges of secondary forest, trackside banks, cultivation, GAH, DTH & FS 18-11, 1 spm.

12

New to São Tomé: 28 Apr. 2008, Chácara, city of São Tomé, photo by RFL (Fig. zz), 1, not kept; 27 Aug. 2016, Monte Café, photo by RFL, 1, not kept; 24 Nov. 2018, by Bom Sucesso Botanical Garden, N0.28829° E6.61193°, 1159 m alt., trackside with herbs near hedges, scrub, wall & cultivation, GAH & DTH et al. 18-1, 1 spm; 27 Nov. 2018, on path above Bom Sucesso Botanical Garden, N0.28° E6.61°, 1159 m alt., on muddy path through herbs and scrub near cultivation, GAH & DTH 18-6F, 1 spm; 10 Dec. 2018, in Bom Sucesso Botanic Garden, N0.2886° E6.6121°, 1156 m alt., botanical garden with groves of trees, lawns & cultivated patches, GAH & DTH 18-22, 1 spm. Identification of the four specimens was confirmed by dissection with all of them having mature male genitalia, despite large differences in body size (length of notum in preserved material 37.5-73 mm). They all showed the very long penial verge tapering to an almost terete point figured by FORCART (1953: pl. 4 fig. 7a) and GOMES (2007: 89 fig. 14), along with other characters set out in the key below. Our material differs from many illustrations of L. alte in having the pale (buff) mid-dorsal line on the notum less conspicuous, but close examination reveals that a narrow and sometimes discontinuous line can be seen along at least part of the posterior half of all four specimens. The species may be a recent arrival on São Tomé and possibly also on Príncipe, the first evidence of its presence being from a photo taken in 2008. Although the type of the species was from India, FORCART (1953: 68) noted that the main development of the genus Laevicaulis is in eastern and central Africa, so he concluded that L. alte was primarily an African species and that its wider distribution (Madagascar, Mascarene Is., India, Sri Lanka, , , Indonesia, New Caledonia, Loyalty Is., ) may have been caused by spread with cultivated plants. Since the 1950s it has been recorded in many other parts of the tropics and subtropics, including Malaysia, Philippines, New Guinea, Fiji, Vanuatu, W. , and Bermuda, being regarded as a pest of cultivated plants in some countries (GOMES & THOMÉ, 2004; COWIE ET AL., 2009).

Pseudoveronicella Germain, 1908 Pseudoveronicella Germain, 1908, Bull. Mus. natn. d'hist. nat., Paris, 14, p. 59; type species Gravieri Germain, 1908 (= V. thomensis Girard, 1893, see below), by monotypy.

Pseudoveronicella forcarti sp. nov. D. Holyoak, G. Holyoak & F. Sinclair (Figs 4C, D)

Type material: Holotype, NHMUK (reg.no. xxxx), spm A in ims (notum length 34 mm) with male genitalia separated, collected 3 July 2019, by FS, A. Pedronho & A. Andrade at type locality (site 2019-07-03-01). Paratypes: Two spm in ims collected with the holotype at the type-locality. Type locality: Príncipe Island, saddle beneath Pico Mesa, N1.58151º E7.35421º, 414 m alt., mid-elevation native forest on exposed ridge. Etymology: The species epiphet honours Lothar H. E. W. Forcart (1902-1990), a zoologist and malacologist from Switzerland, who provided a detailed illustrated monograph of African Veronicellidae (FORCART, 1953) that was fundamental to our work on the family.

Description: Slugs lacking a shell; specimens preserved in ims with notum 30.5-34 mm long, maximum width 10.5-11 mm; body strongly dorso-ventrally flattened throughout, tapering to narrow translucent lateral margins; outline elliptical/narrowly oval, with notum rounded at anterior and posterior ends; upper tentacles projecting a few millimetres beyond anterior end of notum. Preserved specimens have notum light grey, with scattered small and rather indistinct darker spots, the holotype alone having more obvious irregular whitish blotches along midline in posterior one-third of body; surface of notum with even pattern of

13

low rounded tubercles visible at low magnification (×7). Hyponotum pale grey without markings; sole of foot whitish; head pale grey dorsally, whitish at front; upper tentacles grey or grey with white tips and a small black eye-spot centrally near tip. Living slugs not described in detail, but "quite pretty ‒ with greens and yellows and a translucent margin", so natural coloration evidently lost following preservation in ims. Anus a rather large crescentic opening close to posterior end of right hyponotum, not or slightly concealed beneath end of foot. Female pore in lower half of right hyponotum, slightly anterior of middle of (preserved) body length (at 16.5 mm from front of 34 mm holotype, 14 mm from front of 30.5 mm paratype), inconspicuous, or with distinct whiter rim. External pore of male genitalia below and to right of right upper tentacle. Genitalia studied in holotype (Fig. 4C, D) and larger of the two paratypes. Thin-walled penial outer sheath encloses most of penis and distal part of the large penial gland, dividing proximally into separate sheaths enveloping each of these organs, the part covering penial-gland becoming a thick muscular wall proximally. Penis (inside sheath) a short muscular cylinder (less than twice as long as wide), with distinct single annular thickening towards distal end, before narrowing into short, blunt, curved verge almost as long as wider part of penis. Apical portion of verge with two broad blunt apical wings converging towards tip, enclosing a shallow central depression with longitudinal ridges, the opposite side deeply convex. Proximal end of penis with strong retractor muscle attached to outside of penial sheath and passing proximally along body. Distal part of penial gland a long-conical papilla (pointed and probably perforated "stimulator", longer and narrower than verge); middle part longer, muscular, expanding proximally to give clavate structure; large proximal part attached by narrow neck, comprised of group of ca 16 tubular glands, when in situ, running ± parallel to each other, loosely conjoined laterally and somewhat convoluted or curved. Retractor muscle of penial gland arising on proximal end of external sheath of its clavate structure, running proximally to join (larger) penial retractor muscle. Anterior vas deferens starting concealed in body wall close to genital pore, initially thin, then widening to a narrow, long tube, convoluted in situ, passing proximally alongside exterior of penis sheath, ending in centre of proximal end of penis adjacent to attachment of retractor muscle [middle vas deferens not seen; it can be assumed to pass proximally to posterior end of body concealed within the body wall as in other Veronicellidae: GOMES, 2007: 4]. Comments: P. forcarti resembles P. thomensis externally in its small size and green notum, but differs from it markedly in structure of the penial verge (see couplet 3 of Key below; Fig. 4). FORCART (1953: 54-55, pl. 4 fig. 2) recorded that P. (Congoveronicella) schoutedeni Forcart, 1953 has the notum light greenish with indistinct spots, but the verge of that species is peculiar with a wide "spatha" near its tip. All other congeners from the African mainland also differ from P. forcarti in penis structure. The review by FORCART (1953) recognised four subgenera within Pseudoveronicella, the name of one (Hoffmannia Forcart, 1953 non Hoffmannia Heinemann & Wocke, 1877, Lepidoptera) later being replaced by Afroveronicella (FORCART, 1954). Differentiation of those subgenera was based mainly on the form of the penial verge: Vaginina Simroth, 1897 having a fine elongated distal part, the monotypic Congoveronicella Forcart, 1953 having a wide distal spathe, Afroveronicella Forcart, 1954 having the verge "glandiform" (enlarged relative to base of penis), whereas Pseudoveronicella sensu stricto has it "utricular" (narrow relative to base, with P. liberiana (Gould, 1850) placed here). On those criteria, all three species known from São Tomé and Príncipe would be placed in Pseudoveronicella sensu stricto, although both "green" taxa approach subgenus Afroveronicella in having wider verges than P. liberiana. All three specimens of P. forcarti were collected from "tops of saplings (probably all of the same species), at about shoulder height" in native forest on an exposed ridge. Although

14

more observations are needed, the green and yellow coloration of the living slugs and their presence above ground on foliage may imply a partly arboreal micro-, as also suggested below for P. thomensis.

Pseudoveronicella thomensis (Girard, 1893) (Figs 4B, 10A, C)

Veronicella thomensis Girard, 1893, Jornal Sci. math., phys e nat., (2) 3 (10), p. 113; TL Binda, St. Thomé [= Bindá, São Tomé]. syn. Veronicella Gravieri Germain, 1908, Bull. Mus. natn. d'hist. nat., Paris, 14, p. 55; TL São Thomé Island, San Nicolas [probably = São Nicolau]. GIRARD (1893b: 113) gave a short description of a single veronicellid specimen from São Tomé that was in poor condition and not dissected, but he noted the coloration as "uniforme jaune verdâtre" (uniformly greenish-yellow), which in combination with the locality suffices to distinguish it from Pseudoveronicella liberiana and other congeners and validates the name Veronicella thomensis. GERMAIN (1908: 55) named the same species as V. gravieri and gave a very detailed account of the single specimen available to him which was collected by Ch. Gravier in 1906. Germain described (p. 55) the coloration of this taxon as "d'un vert pomme clair", apparently based on a much older account by BRANDT (1835: 250). The monograph by FORCART (1953: 26-36, pl. 3) discounted the name V. thomensis Girard, 1895 as "so insufficient that the species can not be surely recognised" and treated V. gravieri as one of the numerous synonyms of P. liberiana, which was mapped by him as having a wide range in tropical Africa extending from [the former French] Guinea and Liberia eastwards to Uganda and southwards to N. Angola. Dissection of our only fully adult P. thomensis revealed that the apical part of its penial verge was bluntly spatulate (Fig. 4B), whereas several adult and near-adult P. liberiana from São Tomé had a narrower, bluntly pointed end to the verge (Fig. 4A). There are several other reasons for regarding P. thomensis as a distinct species from P. liberiana, in addition to the former being distinctive in its light green coloration dorsally (Fig. 10A, C), whereas the island populations of the latter are light or dull brown dorsally, commonly with dark grey or black blotches and fine dots of blackish-grey (Fig. 10B, D). The first record on São Tomé for P. liberiana was not until the report by GASCOIGNE (1994a: 2), despite it being known on Príncipe since HEYNEMANN (1868: 37, pl. 1 fig. 2) named it from that island as Veronicella myrmecophila. The first report of a green veronicellid from São Tomé in BRANDT (1835) was thus more than 150 years earlier than that for P. liberiana, the late date perhaps indicating that the latter is a recent anthropogenic introduction to this island. Records during 2013-2018 show that P. liberiana is now widespread on São Tomé and that it apparently has a different altitudinal range to that of P. thomensis and they possibly also differ in habitat preferences. Our finds of P. liberiana extend from lowlands near the southern end of the island (68 m alt.) to those of the north-western coastal regions (just S. of Santo António, at ca 24 m, sample 18-14), and upwards to near Monte Café (24 Nov. 2018, near Quinta da Graça, N0.29921° E6.64174°, 678 m alt., scrub, cultivated gardens, roadside banks, few trees, GAH & DTH 18-3, 3 spm). It has been found in secondary forest, but not inside closed forest or in montane forest. In contrast, P. thomensis has been recorded from ca 500 m up to 1380 m. The lowest records were of slugs photographed by RFL in June 2019 (ca 500 m alt.: near stream in secondary forest, above Mirante, São João, Santa Catarina [Lembá district], N0.24364° E6.50242°; ca 530 m alt.: in shade of cocoa plantation, between Java and Abade [Mé Zóchi district], N0.25820° E6.65029°). Otherwise it was recorded only from higher ground, with four records at 700-1093 m alt. in secondary forest and numerous finds at up to 1380 m alt.

15

inside montane forest. Its few records from more open habitats were from inside Bom Sucesso Botanic Garden (at 1156 m) and on a shaded trackside nearby. Whereas P. liberiana was commonly but not exclusively found on the ground, most of the numerous records of P. thomensis were of immature slugs up to 2 m above the ground on leaves of saplings or shrubs, although our only fully adult specimen was from the ground inside montane forest and the slug photographed at ca 530 m was on a dead log. The taxa thus appear to differ in altitudinal ranges (with some overlap between 500 and 678 m alt.), in the habitats occupied, and perhaps in the extent to which they climb up onto vegetation.

Key to species of Veronicellidae of São Tomé and Príncipe

Within the islands identification of species from coloration and locality (island) appears to be reliable, but anatomical characters useful to confirm determinations are also given here.

1. Notum almost uniformly dull grey to blackish-grey, lacking conspicuously darker blotches or fine spots, with narrow and sometimes indistinct buff line along mid-line of posterior half; anus close to mid-line of underside of body, generally concealed by back of foot, rounded with marginal flaps; female opening behind centre of right hyponotum; verge long (> half of penis length), tapering to almost terete point -. Notum (a) green (bright, light, or dull) or (b) light to deep brown with darker blotches and small blackish spots; anus offset from mid-line of underside of body in right hyponotum, more posterior, so not concealed by back of foot, widely lunate; female opening in front of centre of right hyponotum; verge shorter (> half of penis length), towards the point flattened, spatulate or angled, but not terete (Pseudoveronicella spp.) 2

2. Notum light to deep brown with darker blotches, and evenly dotted with small blackish spots which may form papillae; body rather opaque (Fig. 10, B, D); verge short, with narrow canaliculate point (Fig. 4A) P. liberiana -. Notum green, lacking conspicuous dark markings, sometimes with small whitish spots; margins of body translucent; verge flattened and spatulate or short, thick and angular 3.

3. Verge longer; tip of verge wide, spatulate (Fig. 4B); known only on São Tome P. thomensis -. Verge shorter; tip of verge thick and blunt, with two broad blunt angular wings on either side of central depression (Fig. 4C, D); known only from Príncipe P. forcarti sp. nov.

ACHATINIDAE Swainson, 1840 Subfamily Achatininae Archachatina Albers, 1850

Archachatina marginata (Swainson, 1821)

Achatina marginata vary Swainson, 1821, Exotic conchology ..., 1 (1,2), plate with no number; TL apparently not stated, although PILSBRY (1910: 110) attributed Guinea (= West Africa) to Swainson. New to Príncipe: 9 Dec. 2013, just NW. of Cidade de Santo António, N1.6444° E7.4203°, 42 m alt., from base of bank above road, with secondary-forest and cultivation, GAH & DTH ST9, 2 sh (from ca 15 seen).

16

The species is native in continental West Africa where it is widespread (e.g. PILSBRY, 1904: 109, 1905: pl. 24 figs 22, 23, pl. 25 fig. 26; SCHILEYKO, 1999: 484; COSSIGNANI, 2014: 156- 158). It was not found on São Tomé or Príncipe by collectors in the nineteenth- or early twentieth-centuries. GASCOIGNE (1994a: 3-4) cited anecdotal evidence that it had been introduced to São Tomé within the preceding 30-40 years and gave a map of its range, which was then restricted to the lower elevations in the north and centre of that island. In 2013 we also found it in the extreme south of the island. A detailed study of its colonisation has been carried out since then by PANISI (2017). Our record from Príncipe in 2013 confirmed that it had also been introduced and become established there. In 2018 it was clearly widespread on the island and well known to local people. Accumulations of scores of shells of snails collected for food by local people were seen by us in native forest (e.g. 4 Dec. 2018, along path to Santa Joaquina, N1.61313° E7.39683°, at 240 m alt.).

Limicolaria Schumacher, 1817

Limicolaria flammea (O.F. Müller, 1774)

Helix flammea O.F. Müller, 1774, Verm. terr. fluv., 2, p. 87; TL "Senegal" (fide Adanson) & "In Guinea ad arcem Cristiansburgensem". New to São Tomé: Nov. 2018, around Mesquita, Água Grande District, plentiful in gardens, leg. MP, 2 sh (in plantations and found to be present in gardens all along the main street that goes from Mesquita up to the border of Mé-Zochi District); 9 Nov. 2018, garden outside the National Herbarium, Boa Nova, Alto Potó [Mé Zóchi district], N0.33678° E6.65057°, ca 500 m alt., RFL; 28 Nov. 2018, Mesquita, N0.34299° E6.70533°, 69 m alt., waste-ground around industrial building, among low weedy herbs, with bananas planted along edge, DTH, GAH, RFL & MP 18-8, 3 sh & bod, 7 spm; 2 Oct. 2019, Água Crioula, Mé-Zochi District, in shade plantation, RFL. Native of continental Africa, known from Sierra Leone, Guinea, Nigeria, Cameroon and Zaïre (CROWLEY & PAIN, 1970; SCHILEYKO, 1999: 474; COSSIGNANI, 2014: 165-166). Broader shells figured by COSSIGNANI (2014: 166) from Tanzania are only tentatively placed as this species. L. flammea has been known as an introduction in Singapore since 2006 (TAN & CLEMENTS, 2011; TAN & LOW, 2011), where it is potentially damaging to the valuable horticultural industry. Local people in São Tomé date its introduction in Água Grande District to around three to five years ago. They complain that it causes damage to cultivated plants and in 2019 some were using baits of jackfruits (Artocarpus heterophyllus) combined with pesticides to remove the species. However, others are deliberately spreading L. flammea in the mistaken belief that they are juveniles of the edible Archachatina marginata. On 28 Nov. 2018 (see above) we found L. flammea living at high density accompanied by smaller numbers of A. marginata.

Subfamily Petriolinae Schileyko, 1999

Petriola Dall, 1905

DALL (1905: 143) introduced the generic name Petriola with type species Achatina marmorea Reeve, 1850 (by original designation), as a replacement name for Trichodina Ancey, 1888 which is preoccupied by Trichodina Ehrenberg, 1831 (Ciliophora). PILSBRY (1905) and GERMAIN (1916) treated these large subulinids which are endemic on São Tomé as forming the subgenus Petriola within Bocageia Girard, 1893. However, this seems to be

17

unwarranted since the type of Bocageia is Bulimus lotophagus Morelet, 1848 an endemic of Príncipe, having a shell with fewer whorls than the São Tomé species and a very shallow suture (BREURE ET AL., 2018: 330 fig. 605); its genital anatomy is unknown. The São Tomé species had also been placed in Homorus Albers, 1850 (by CROSSE, 1888a; NOBRE, 1891; GIRARD, 1893a, 1893b), but that genus has narrowly-conical rather than subcylindrical shells, with a proportionately longer mouth which is pointed rather than rounded below. Also, the scanty information on its genital anatomy (PILSBRY, 1919: 114; SCHILEYKO, 1999: 522-523, fig. 681) shows it differs in having the free oviduct a little longer than the vagina, whereas the free oviduct is "very short" in Petriola clava (SCHILEYKO, 1999: 521-522, fig. 679) or absent (our Fig. 5D of P. marmorea and 5H and I of P. monticola). Other anatomical features of Petriola as a whole are the penis which can be extruded from genital pore, containing a short distal verge (at least in P. clava and P. marmorea); well-defined epiphallus; penial retractor muscle attached to epiphallus/vas deferens junction [contra SCHILEYKO, 1999: 522-523, fig. 681], the muscle rather short and descending to join columellar muscles; short genital atrium; rather long vagina; small bursa copulatrix on very short duct, the reservoir lying alongside distal end of spermoviduct; spermoviduct large and long; some mature individuals with 1-5 embryos developing shells inside oviducal part of spermoviduct; common hermaphrodite duct long, narrow, convoluted in distal portion, almost straight in proximal portion. Hence, we retain the São Tomé species in the endemic genus Petriola for now, although more extensive research on genital anatomy and molecular- sequence data for Bocageia lotophaga and the numerous poorly known nominal taxa from the African mainland may lead to changes in the future. SCHILEYKO (1999: 520) named the subfamily Petriolinae within the Subulinidae mainly on the basis of the structure of the penis, since "no one other subfamily has such conspicuous organization of copulatory apparatus", presumably referring to possession of a distinct epiphallus and a verge inside the distal penis. Molecular-phylogenetic data presented by FONTANILLA ET AL. (2017: fig. 1) for "Bocageia sp." gave additional justification for recognition of Subulinidae (Petriolinae) as a subfamily, which became Achatinidae subfamily Petriolinae in BOUCHET ET AL. (2017), but the "Bocageia sp." which provided the only sequence data involved was a specimen from São Tomé corresponding to Petriola sp. SCHILEYKO (1999: 521-522, fig. 679) published the only detailed account of the genital anatomy of any of the Petriola species, based on P. clava. Besides revising that description (using 11 specimens from 4 populations), we have obtained the first data on the genital anatomy (Fig. 5) of other congeneric species from São Tomé recognised by PILSBRY (1905) and on Achatina (Subulina) costulata Greeff, 1882 (GREEFF, 1882: 519). These fall into three natural groups according to length of the penis and epiphallus, but they all show a general similarity, so we retain them in the same genus. Along with study of numerous shell samples (see below), this leads us to recognise three species in Petriola which may be distinguished as follows:

1. Periostracum dark brown to blackish, in fresh shells with narrow axial riblets and higher lamellae bearing spines, intersecting spiral riblets to give decussate pattern; shell broad (B >11.0 mm in mature snails; breadth 5 mm below apex >4.4 mm); penis shorter than vagina (Fig. 5D-F, I); small section of vas deferens passing inside outer sheath of penis (Fig. 5I) P. marmorea (Reeve, 1850) -. Periostracum brown to whitish, smooth or bearing regular or irregular axial riblets or ribs, but without spines or decussate pattern; shell narrower (B <11.5 mm in mature snails; breadth 5 mm below apex <5.5 mm); penis shorter or longer than vagina; all of vas deferens passing outside outer sheath of penis (Fig. 5) 2

18

2. Shell wider (B >7.7 mm in mature snails; breadth 5 mm below apex 3.9-5.4 mm); body- whorl almost smooth or with wavy axial lines; periostracum of fresh shells brown, glossy; proximal part of distal portion of epiphallus not slender, not markedly convoluted in situ (Fig. 5A) P. clava (L. Pfeiffer, 1864) -. Shell narrower (B <7.0 mm in mature snails; breadth 5 mm below apex <3.9 mm); body- whorl smooth, with axial riblets or ± strong axial ribs; periostracum of fresh shells, whitish to pale brown; proximal part of distal portion of epiphallus slender, convoluted in situ (Fig. 5H) P. monticola (Morelet, 1866)

PILSBRY (1905: 184, pl. 56 figs 13-15) treated Homorus massonianus Crosse, 1888 (CROSSE, 1888a: 22, pl. 1 fig. 3) as a valid species, but GERMAIN (1916: 264) was probably correct in suggesting it may be a variety of P. marmorea with less developed shell sculpture, since Crosse gave the maximum diameter of the shell as 13 mm. GERMAIN (1916: 163, 168, 264) also emended the species epiphet to Bocageia (Petriola) massoni Crosse, but the change was unnecessary. COSSIGNANI (2014: 195) figured two shells as Bocageia massoni (Crosse, 1868 [sic]), but they are clearly P. marmorea because they show coarsely decussate periostracal sculpture with some small spines visible. A syntype of Achatina monticola Morelet, 1867 is figured by BREURE ET AL. (2018: 358 fig. 719). Although immature, this shell clearly shows axial ribs well developed beneath the suture on the spire but weak or absent in the middle of each whorl. GREEFF (1882: 519) named a strongly ribbed form as Achatina (Subulina) costulata and another with weaker ribs as Achatina (Subulina) subcrenata, the latter being treated as a synonym of P. monticola by PILSBRY (1905: 187) and GERMAIN (1908: 62). Stenogyra costulata Greeff was retained as a species by CROSSE (1888: 25) and GIRARD (1893a: 33), but it was reduced to Homorus monticola var. costulata Greeff by GIRARD (1893b: 96) and also treated at varietal rank by PILSBRY (1905: 187) and GERMAIN (1916: 156). GERMAIN (1916: 269-271) named an additional variety as Bocageia (Petriola) monticola Morelet var. sculptisuturata, with specimens from "Ribeira Palma, entre 400 et 700 mètres d'altitude, Ribeira Palma, dans la forêt, entre 500 et 600 mètres d'altitude and Vista Alegre, entre 200 et 300 mètres d'altitude". However, his new variety occurred with typical shells at the first and last of those localities and its characters seem to be similar to those of the syntype we mention above. Although the type locality of P. montana was in the hills ("ad 2,500 ped. altitudinis" = 762 m: MORELET, 1866: 161) and we have collected it up to 1416 m alt. in low montane forest near Lagoa Amélia (our sample 18-10), we can also confirm reports that it occurs down to near sea-level (0-300 m: GERMAIN, 1916: 270), so the species epiphet is not always appropriate. Our own material includes a majority of samples with smooth shells lacking ribs or with them only slightly developed, two samples with strong axial ribs continuous across the whorls, along with others having weak ribs or ribbing only just below the suture. Shells from a wet basalt cliff at ca 6 m alt. near the shore N. of Santa Catarina (sample ST6) have striking strong almost regular axial ribs. The only shells we have found with equally strong (but less regular) ribbing are from the upper altitudinal extreme in sample 18-10 from near Lagoa Amélia and the latter locality, also produced nearly smooth shells from close by along with a few intermediates. Dissection of a strongly ribbed adult individual from 18-10 did not disclose any differences in genital anatomy from smooth-shelled individuals from lower altitudes in montane forest. The lack of any anatomical difference and the occurrence of populations at other localities with ribbing developed to intermediate extents leads us to regard all these forms as conspecific.

Subfamily Opeatinae Thiele, 1931

19

Pseudopeas S. Putzeys, 1899

Pseudopeas crossei (Girard, 1893)

Opeas Crossei Girard, 1893(b), J. Sci. math., phys. e nat., Lisboa, ser. 2, 3, p. 105, pl. 1, fig. 13; TL Príncipe "Ò que S. João á une altitude de 200 m." New to São Tomé: 10 Dec. 2018, on path between Bom Sucesso and Macambrará, near Rio Manuel Jorge, N0.28571° E6.60912°, 1114 m alt., horticulture areas with bananas, groves of scrub & trees alongside track, GAH 18-23, 1 sh (live-collected). A species previously known from Príncipe (GIRARD, 1893b: 105, pl. 1, fig. 13; 1894: 201; PILSBRY, 1906: 115, pl. 15, figs 60, 61; GERMAIN, 1912b: 340; 1916: 167) and Bioko (GASCOIGNE, 1994a: 2, 5; WRONSKI ET AL., 2014: 164, Table 1).

Subfamily undet. Genus Nothapalus von Martens, 1897

Nothapalus solitarius sp. nov. G. Holyoak & D. Holyoak (Fig. 2A)

Type material: Named and described only from the unique holotype shell: NHMUK reg.no., collected 4 Dec. 2013, GAH ST4. Type locality: São Tomé Island, path to Lagoa Amélia from Bom Sucesso Botanical Garden, N0.2786° E6.6000°, ca 1300 m alt., montane forest on slope with understorey of saplings. Etymology: The species epiphet uses the Latin solitarius (meaning alone, or by itself) referring to the single shell specimen found despite repeated visits, and its distinctness from other species.

Description: Shell dextral, H 4.69 mm, B 2.40 mm, H/B 1.96; body-whorl height 3.29 mm (70.1% of H); AH 2.06 mm (43.9% of H), AB 1.03 mm, AH/AB 2.00; whorls 3.9. Shell shape rather narrowly conical, with rounded base. Whorls of spire slightly rounded, with definite shoulder close below suture; suture somewhat oblique, impressed but shallow, appearing as narrow double line. Body-whorl large, approximately oval. Aperture ear-shaped with rounded lower palatal margin, gently curved outer palatal margin, curved parietal margin; columellar margin almost straight in upper half sharply truncate just below middle, the lower half with strongly curved shallow recess. Peristome (apparently not fully mature) thin, plane, with membranous fringe except on columellar margin, which widely reflected in upper half, overlapping but not closing umbilicus. Umbilicus a narrow deep chink, the interior visible only in ventral view. Protoconch small, slightly domed, with faint granular microsculpture. Teleoconch with low, rather irregular axial ribs that are slightly curved on upper whorls and more strongly curved on body-whorl; spiral microsculpture (of thin grooves) crossing the low ribs, rather indistinct on body-whorl, more conspicuous on whorls 2 and 3. Shell thin, translucent, whitish with weak yellow tinge, moderately glossy. Body (dried inside shell) and anatomy unknown. Comments: Because the anatomy is unknown, allocation to a genus is tentative. It appears to be a species of Subulinidae from the thin, elongate, whitish shell, axial ribbing and lack of apertural barriers, and this family also seems likely in view of its wide radiation and prolific speciation on São Tomé and Príncipe. The provisional assignment to Nothapalus, represented by several species in the Congo basin, is based on the narrowly conical/ovoid shell outline, few whorls, flattened whorl profiles and especially, the markedly truncate (cf. PILSBRY, 1919: 129-134, pl. 19 figs 1-6). Nevertheless, Nothapalus differs in its larger shells (H 13-22.3 mm), closed umbilicus and lack of spiral shell sculpture. Some species of the

20

genus Curvella Chaper, 1885 are similar in having few whorls, axial ribbing, whitish shells and also rather small size (H 6 mm in some South African species: HERBERT & KILBURN, 2004: 136), but none of them have the columella truncated. Outside the Subulinidae, Micractaeon shows a conspicuous similarity in shell form (VAN BRUGGEN & DE WINTER, 1995), but it differs in having a relatively tiny shell (H up to 2.1 mm), brown shell coloration and spiral sculpture on the shell consisting of distinctive lines of rounded pits (clearly visible at ×56 with strong oblique illumination using our microscopes, whereas Nothopalus solitarius shows faint spiral grooves and continuous lines not rows of pits). Some genera of the family Ferussaciidae also have shells of similar general form, but the family is not known in tropical west Africa. Thus, more material is needed to use anatomical or molecular characters to establish the affinities of this species.

MICRACTAEONIDAE Schileyko, 1999 Micractaeon Verdcourt, 1993

Micractaeon koptawelilense (Germain, 1934)

Pseudopeas koptawelilensis Germain, 1934, Bull. Mus. Natn. Hist. nat., Paris, (2) 6, p. 380; TL , "Mount Elgon, vallée de la Koptawelil, à 2.300 mètres d'altitude, au nord de la rivière Koitoboss". VAN BRUGGEN & DE WINTER (1995: 80) pointed out that the correct orthography of the name would be Pseudopeas koptawelilense. New to São Tomé: 24 Nov. 2018, Morro Claudina (near Bom Sucesso), N0.2900° E6.6055°, 1257 m alt., montane rainforest (sievings from shelly debris at thrush anvil), GAH, DTH et al. 18-2C, 5 sh; 27 Nov. 2018, Morro Claudina, N0.28909° E6.60442°, ca 1290 m alt., montane forest with understorey (sievings from shelly debris at thrush anvil), GAH & DTH 18-6A, 5 sh; ditto, N0.29181° E6.60592°, 1288 m alt., floor of montane forest near tiny clearing (sieved from shelly debris at thrush anvil among buttress roots of tree), sample 18- 6C, 3 sh; ditto, N0.29036° E6.60537°, 1254 m alt., floor of rather open montane forest with some alien trees (sievings from shelly debris at thrush anvil), sample 18-6D, 2 sh; ditto, N0.29011° E6.60537°, 1255 m alt., floor of montane forest with some alien trees, near edge of recent clearing (sieved from shelly debris at thrush anvil), sample 18-6E, 2 sh; 28 Nov. 2018, Cascata de São Nicolau, N0.2857° E6.6257°, 885 m alt., around waterfall on steep rocky slopes with secondary forest, GAH, DTH et al. 18-7, 1 sh. VAN BRUGGEN & DE WINTER (1995) provided good illustrations and revised the nomenclature, taxonomy and distributional data on this species, showing that it has a wide range in tropical Africa from Ghana and Cameroon south-eastwards to western Kenya, south- eastern Zaïre and Malawi. Although recorded from Bioko (WRONSKI ET AL., 2014: 164, Table 1) there were no records hitherto from oceanic islands offshore of Africa.

STREPTAXIDAE J.E. Gray, 1860 Subfamily Enneinae Bourguignat, 1883

Gulella L. Pfeiffer, 1856

Gulella azeitonae sp. nov. D. Holyoak, G. Holyoak & F. Sinclair (Fig. 2B)

Type material: Described only from the unique holotype, NHMUK reg.no, sh & (incomplete) bod, collected 3 May 2019, by A. Pedronho & R. Fernandes, 2019-05-03-02. Type locality: Príncipe Island, Floresta de Azeitona, N1.64968º E7.39879º, 194 m alt., tall secondary forest.

21

Etymology: The species epiphet azeitonae is based on the type-locality in the Floresta de Azeitona.

Description: Shell dextral, H 2.90 mm, B 1.74 mm, H/B 1.66; AH 0.73 mm, AB 0.67 mm, AH/AB 1.09; whorls 6.9. Other measurements (based on those used by DE WINTER & VASTENHOUT, 2013: 608 fig. 1): body-whorl height 1.46 mm, peristome height 0.93 mm, peristome width 0.91 mm, width across five successive rib-intervals and six ribs (RD) in middle of penultimate-whorl 1.46 mm). Shell shape biconical, with apex somewhat rounded and base of shell broadly rounded. Whorls increasing gradually in width and descending gradually up to start of body-whorl; most of body-whorl narrower than penultimate-whorl and not descending. Whorls of spire rounded in profile; suture moderately deep. Body-whorl somewhat constricted behind aperture. Middle of exterior of last half of body-whorl with prominent external groove just below mid-height that disappears towards aperture. Aperture asymmetrically heart-shaped, with broadly rounded base, rounded sinus around parietal- upper palatal junction and somewhat straighter columellar and parietal margins. Peristome continuous, reflected on palatal and columellar sides of aperture, not reflected on parietal side; section above parietal lamella with flap raised away from aperture beyond line of rest of peristome. Two apertural barriers clearly visible: a long low parietal lamella with serrated crest running into aperture from line of peristome, and a rounded mid-palatal projection inwards of the unthickened peristome coresponding to external hollow. External groove on body-whorl apparently corresponding to deep-set mid-palatal lamella (dry body prevents clear view from aperture). Protoconch and start of teleoconch (up to whorl 2.5) almost smooth, with granular or somewhat decussate microsculpture (lacking clear spiral elements. Rest of teleoconch with strong sculpture of rather evenly spaced axial ribs (ca 30 counted around underside of body-whorl), the ribs nearly straight to slightly curved, mainly with slight transverse orientation. Interstices between ribs smooth (no trace of spiral microsculpture). Protoconch, early whorls of teleoconch, peristome and interior of aperture whitish; rest of teleoconch pale brownish. Shell not thick, slightly translucent; somewhat glossy, especially on crests of axial ribs. Exterior of body (described from visible parts pulled from specimen that had been drowned in water then kept in 80% ethanol for four months): sole of foot white; exterior of body orange; eye-spot in upper tentacle black; dorsum of mantle from inside shell orange. Genital anatomy unknown; preserved foreparts of body may contain the radula but they are too incomplete for study of the genitalia. Discussion: The new species differs strikingly from the three species of Gulella hitherto known from Príncipe (G. crystallum Morelet, 1848, see BREURE ET AL., 2018: 262 fig. 305 for syntype; G. sorghum Morelet, 1848, see BREURE ET AL., 2018: 435 fig. 1064 for syntype; G. joubini (Germain, 1912), see SCHILEYKO, 2000: 812 fig. 1062 for shell figure) in possessing strong axial ribs on the shell and in the number or position of its apertural barriers. Unlike the three Gulella species already known from Príncipe, G. azeitonae sp. nov. somewhat resembles the segregate genera Avakubia Pilsbry, 1919 and Pseudavakubia de Winter & Vastenhout, 2013, in its combination of small shell size, biconvex shell profile, strong axial ribs, few apertural barriers, as well as having a range in west or tropical Africa. However, comparisons of the unique holotype with the detailed monograph of those genera by DE WINTER & VASTENHOUT (2013) reveal numerous differences: it lacks spiral sculpture on the protoconch and in the interstices between ribs on the teleoconch; the body-whorl narrows towards the aperture; the back of the body-whorl has an external groove; the visible part of the parietal lamella has a serrated rather than smooth edge; the peristome has a prominent flap projecting outwards over the aperture; the umbilicus is closed (as in

22

Pseudavakubia, but unlike Avakubia; but differing from Pseudavakubia and resembling Avakubia more in the regular increase of early whorls of the spire). Because of these differences, we prefer to place the new species in Gulella, which remains a "catch-all" genus containing disparate elements. Much more study of this largest of African land-snail genera is needed to resolve its internal relationships and to place the four endemic Príncipe species in a wider context, although ROWSON & HERBERT (2016) have made a valuable start.

Streptostele Dohrn, 1866

DOHRN (1866b: 128-133) treated three species that had formerly been placed in the genus Bulimus in his new genus Streptostele, successively as S. lotophaga, S. fastigiata and S. folini, but did not designate a type species. He stated that he had not seen Bulimus lotophagus, which he did not describe, whereas he gave detailed descriptions of the other two species and his description of the new genus Streptostele was clearly based on those and not on B. lotophagus. GIRARD (1893b: 102) cited several authors at length who had pointed out that B. lotophagus was only distantly related to S. fastigiata and S. folini, leading to his comment: "le Bulimus lotophagus ne peut être considéré comme le type du genre Streptostele." Hence, GIRARD (1893b: 100) introduced the new genus Bocageia with B. lotophaga as its type (by monotypy), but he did not fix the type of Streptostele. Despite this, C.F. ANCEY (in VIGNON, 1888: 68 footnote) designated Bulimus lotophagus Morelet, 1848 as type species of Streptostele, without any explanation, apparently in ignorance of previously published work. PILSBRY (1919: 181 footnote) stated: "I am not accepting Ancey's name-juggling because it appears to me that he erred in selecting as type of Dohrn's group a species which that author expressly states that he had not seen, and which disagrees in conspicuous and essential characters with Dohrn's diagnosis. It was really a species inquirenda from his standpoint, and as such was not available as a type, under existing rules. This is a case which should be decided by the International Commission on Zooological Nomenclature." The ICZN Code Art. 67.2.5 currently (2019) states: "A nominal species is deemed not to be originally included if it was doubtfully or conditionally included, or was cited as a species inquirenda, or as a species incertae sedis." We therefore follow PILSBRY (1919) and established contemporary usage (e.g. SCHILEYKO, 2000: 802-806) in regarding the first valid fixation of the type species of Streptostele as being that by E.A. SMITH (1890: 96) who selected Bulimus fastigiatus Morelet, 1848 (by subsequent designation).

Streptostele abbreviata sp. nov. D. Holyoak, G. Holyoak & F. Sinclair (Figs 3A, 6A-C)

Type material: Described only from the unique holotype, NHMUK reg.no, sh & bod, collected 3 July 2019, by FS, A. Pedronho & A. Andrade, site 2019-07-03-01. Type locality: Príncipe Island, saddle beneath Pico Mesa, N1.58151º E7.35421º, 414 m alt., mid-elevation native forest on exposed ridge. Etymology: The generic name Streptostele has been treated as a feminine noun. The species epiphet is from abbreviatus (Latin, meaning shortened), because its shell shape is much shorter than in the allied S. fastigiata (Morelet, 1848).

Description: Shell H 12.29 mm, B 5.89 mm, H/B 0.46; AH 4.87 mm, AB 2.84 mm, AH/AB 1.71; maximum height of body-whorl (in apertural view) 7.7 mm (60.2% of H). Shell dextral, with 5.7 whorls; outline narrowly-ovate with conical spire. Upper whorls rounded, body whorl with flatter outline; suture moderately deep. Aperture with rounded base, columellar and palatal margins almost parallel and both approximately straight, long

23

almost straight parietal margin and rounded sinus near junction of palatal and parietal regions. Peristome slightly thickened, reflected on lower-palatal and columellar margins; a thin callus extending from reflected columellar lip across parietal area. Umbilicus lacking. Protoconch small and rounded with spiral microsculpture that continues onto third whorl (teleoconch) where longitudinal riblets begin. Rest of teleoconch with strong low longitudinal ribs; also faint traces of spiral microsculpture on parts of body-whorl. Shell thin, somewhat translucent, moderately glossy; pale cream, shading to whitish on peristome and base of body-whorl. Body (and anatomy) were described from specimen drowned in water then preserved in 80% ethanol before being pulled from shell about three weeks later. This body was partly contracted, with head withdrawn under mantle-collar. During dissection coloration was noted: skin on external foreparts and front of flanks white, tail-tip white and back of sole of foot white. The mantle-collar and surface of mantle inside shell were also mainly white, but a wide band across front half of dorsum of mantle was bright yellow. Two other small areas on dorsum of mantle (corresponding to position of retracted upper tentacles inside the body- wall) each formed a yellow streak containing a small black eye-spot. Upper three whorls of spire containing digestive gland, pale brown with weak network pattern of thin darker lines (in intricate pattern suggestive of alveoli of mammalian lungs). As in dissections of the congeners S. fastigiata and S. folini (Morelet, 1848), the interior of the front of the body was dominated by a very large muscular buccal mass continuing proximally as a large and muscular-walled oesophagus. The distal genitalia were small compared to the digestive organs, with the female tract conspicuously more slender than the male tract. Genital pore on right flank of forepart of body. Genital atrium a short and rather wide cylinder, dividing proximally into distal penis and distal vagina. Penis a large, long organ, muscular inside a firm sheath, widening proximally to give clavate, curved outline. Proximal part of penis forked (Fig. 6A, B), to give larger penial caecum and smaller, shorter proximal penis, each with a separate retractor muscle arising apically before the muscles fuse into a flat fan of muscle which becomes narrower as it continues proximally as the combined penial retractor muscle to join other muscles along columellar axis of the body in mid-spire. Epiphallus and flagellum lacking. Vas deferens a thin tube, starting near distal end of spermoviduct, passing distally alongside free oviduct into the angle formed with distal part of penis, then passing proximally alongside penis, without entering penis sheath until it does so almost at tip of proximal penis (adjacent to insertion of its retractor muscle). Vagina very short, its precise point of separation from distal end of duct of bursa copulatrix unclear since both organs thin-walled, flattened in situ, bound side by side with connective tissue and too delicate to separate. Distal part of free oviduct narrow, convoluted into several folds, a thin- walled (translucent), flattened tube about half as long as penis overall. Duct of bursa copulatrix rather wide but thin-walled and flattened in distal part alongside free oviduct; its proximal part a very thin long tube, when in situ following course proximally along distal half of spermoviduct, then diverging widely to cross externally over the oesophagus, returning back over it further proximally, with bursa reservoir lying partly on spermoviduct. Reservoir oval, tapering into duct, thin-walled. Spermoviduct long, flattened tapering at proximal end without passing into a clearly defined albumen gland. Peculiar cylindrical sac ca 1.5 mm long (Fig. 6A, C) resting against proximal end of spermoviduct is whitish, opaque and firm-walled, connected by folded duct that becomes thin as far as it could be traced proximally; the duct with a distinct thin strand passing proximally that may represent retractor muscle. Function and hence terminology of the sac unclear: possibly a very long talon, possibly a seminal vesicle.

24

Comments: The two large endemic species of Streptostele hitherto known from Príncipe both have elongate, very narrowly conical shells, S. fastigiata being conspicuously larger (H 14-29, mostly 18-19 mm, B 6-7 mm: MORELET, 1848: 352-353, 1858), S. folini usually smaller and always narrower (H 17 mm, B 5 mm: MORELET, 1858: 16-17; H 16 mm, B 4.5 mm: DOHRN, 1866b: 131). BREURE ET AL. (2018: 284 fig. 399, 287 fig. 415) provided good illustrations of Morelet's syntypes of both species. In our experience, the narrower apical whorls of S. folini allow immature as well as adult individuals to be separated easily by direct comparisons. S. abbreviata sp. nov. differs from both in having a much shorter shell when adult (holotype H 12.29 mm, B 5.89 mm) with a narrowly-ovate rather than very narrowly conical outline; the mature holotype has 5.7 whorls compared to 8-10 whorls in mature individuals of the other species; its body whorl forms 60% of shell height, compared to much less than 50% in the other species. Indeed, it differs from them so conspicuously in shape that our first impressions suggested it might be a white-shelled specimen of . Our field collections show that both S. fastigiata (11 sites) and S. folini (7 sites) are widespread in wooded habitats on the island, with both recorded from plantations, secondary-forest and native forests, often living together in ground-litter. Whereas S. fastigiata had a wide altitudinal range recorded (24-860 m) including low mossy forest in the mountains, S. folini was found only at 24-194 m. The only record for S. abbreviata was from 414 m alt., in mid-elevation native forest on an exposed ridge. Body coloration in living individuals of S. fastigiata and S. folini is conspicuously pink to almost orange, as in the other common streptaxid, Gulella crystallum (Morelet, 1848), which commonly coexists with both of these Streptostele species. The body colour of living S. abbreviata remains to be confirmed, since pink and orange colours of the other streptaxids disappear quickly after preservation in alcohol. SCHILEYKO (1999: 521 fig. 680) figured a shell of the new species misidentified as 'Bocageia lotophaga (Morelet, 1848). "Ile du Prince". Paris' His figure clearly differs from photos of a syntype of Bulimus lotophagus figured by BREURE ET AL. (2018: 330 fig. 605) in the narrower spire with deeper sutures and prominent longitudinal ribs and closely resembles the holotype of S. abbreviata (our Fig. 3A). Dissections of specimens from Príncipe of Streptostele fastigiata (two snails from different populations), S. folini (one) and S. abbreviata sp. nov. (holotype) showed they have generally similar genital anatomy (Fig. 6), so they may represent an endemic radiation on Príncipe island. Taking account of different degrees of maturity among the specimens dissected, the main anatomical difference among these species is that S. folini has the penial retractor muscle undivided at its distal end and attached only to the proximal end of its penial caecum (Fig. 6E), whereas S. fastigiata (Fig. 6D) and S. abbreviata (Fig. 6A, C) have a divided distal end to the penial retractor consisting of a larger portion attached to the penial caecum and a smaller portion attached to the apex of the penis itself (close to the insertion of the vas deferens). The difference in the retractor muscle is rather clearly associated with the smaller size of the free part of the proximal penis in S. folini than in the other two species. If in future separate generic (or subgeneric) rank is thought useful for S. folini, it is the type species by monotypy of Campylaxis Ancey in Vignon, 1888 (March), Bull Soc. Malac. France, 5, p. 68 footnote. ANCEY (IN VIGNON, 1888: 68) wrote "Ancey 1885" for his new name, but apparently in error since although ANCEY, 1885: 143-144 discussed Streptostele, including folini, and named the new genus Tomostele, he did not use the name Campylaxis.

Tomostele Ancey, 1885

Tomostele musaecola (Morelet, 1860)

25

Achatina musaecola Morelet, 1860, J. Conchyl., 8, p. 190; TL "Guineam". BREURE, AUDIBERT & ABLETT (2018: 361) figured one of three syntypes in NHMUK and its label in Morelet's handwriting giving the locality as "Gabon", so we herewith restrict the type- locality to Gabon. New to São Tomé: 6 Dec. 2013, S. of EN2 and ca 0.4 km W. of Vila Aida, N0.0897° E6.5903°, ca 181 m alt., tall forest on ridge & slopes with understorey of saplings, GAH ST7, 1 sh (live-collected); 11 Dec. 2013, by EN2 ca 1.2 km N. of Monte Mário "roça", N0.0694° E6.5522°, ca 101 m alt., roadside bank with ferns & herbs at foot of slope with secondary forest, GAH & DTH ST11, 1 sh (live-collected). HAUSDORF & BERMÚDEZ (2003) found that Luntia insignis E.A. Smith, 1898 from the Caribbean island of Trinidad was a synonym of Streptostele (Tomostele) musaecola (Morelet, 1860). It had been widely reported under the name L. insignis as an introduction in tropical and subtropical America (Bermuda, Nicaragua to Colombia, Lesser Antilles, , Surinam). GERBER & CLARK (2015) reported the first record of S. musaecola becoming established in the continental U.S.A. (from ) and also cited published reports of it from Melanesia (Vanuatu, Fijian Is., American Samoa) and Polynesia (Cook Is., Society Is.). COWIE (1998) cited an unpublished report by Miller that this carnivorous alien species might have been implicated in the extinction of a native species in American Samoa. Tomostele Ancey (ANCEY, 1885: 43) has Achatina musaecola Morelet, 1860 as type by original designation (albeit as Achatina muscola [sic], following PFEIFFER, 1868, Mon. Helic. viv., 6: 239). We treat Tomostele at generic rank rather than as a subgenus of Streptostele because it is a much smaller shell, differing from the Príncipe taxa also in having the base of the columella markedly truncate; its genital anatomy has apparently not been described. We adopt the generic name Tomostele here for T. musaecola, but leave it to others to seek appropriate generic placements for numerous other species from the African mainland and Malagasy region currently lumped together in Streptostele. This task is perhaps best left until anatomical or molecular data is obtained for them, since placements based on shell characters alone seem unlikely to be reliable.

PUNCTIDAE Morse, 1864 Punctum Morse, 1864

Punctum camerunense de Winter, 2017

Punctum camerunense de Winter, 2017, Arch. Molluskenk., 146, pp. 204-206, fig. 1A-I, table 1; TL Cameroon, South West Region, ca 7 km SW. of Buea; undisturbed cloud forest; 04°08'N., 009°11'E; 1300-1400 m a.s.l. New to São Tomé: 27 Nov. 2018, Morro Claudina (near Bom Sucesso), N0.28909° E6.60442°, ca 1290 m alt., floor of montane forest with understorey (sieved from shelly debris at thrush anvil), GAH & DTH 18-6A, 5 sh; ditto, N0.2890° E6.6044°, sieved from ground-litter on floor of montane forest, sample 18-6B, 2 sh; ditto, N0.29181° E6.60592°, 1288 m alt., floor of montane forest near tiny clearing (sieved from shelly debris at thrush anvil among buttress roots of tree), sample 18-6C, 1 sh; ditto, N0.29011° E6.60537°, 1255 m alt., floor of montane forest nwith some alien trees, near edge of recent clearing (sieved from shelly debris at thrush anvil), sample 18-6E, 1 sh; 9 Dec. 2018, path from above Bom Sucesso Botanical Garden towards Lagoa Amélia, ca N0.2823° E6.5964°, 1290-1415 m alt., montane forest (sieved), GAH & DTH 18-21, 2 sh. DE WINTER (2017) described this species from four shells found at two adjacent localities in the Mount Cameroon complex (South West Region, Cameroon), in forest at 1300-1400 m alt. and it has not hitherto been reported elsewhere.

26

CHAROPIDAE Hutton, 1884 Subfamily Charopinae Trachycystis Pilsbry, 1893

Trachycystis iredalei Preston, 1912

Trachycystis iredalei Preston, 1912, Proc. zool. Soc. London, 1912, pp. 187-188, pl. 32 fig. 8; TL Between the Igembi Hills and Nyeri, British East Africa. New to São Tomé: 24 Nov. 2018, São Tomé Island, Morro Claudina (near Bom Sucesso), N0.2900° E6.6055°, 1257 m alt., on ground in montane rainforest (sieved from shelly debris of thrush anvil), DTH, GAH, et al. 18-2C, 1 sh; 27 Nov. 2018, Morro Claudina, N0.29181° E6.60592°, 1288 m alt., on ground among buttress-roots of tree near tiny clearing in montane rainforest (sieved from shelly debris of thrush anvil), GAH & DTH 18-6C, 1 sh. Description: Site 18-2C, shell: H 0.78 mm, B 1.77 mm, H/B 0.44; AH >0.56 mm, AB >0.64 mm; UB 0.66 mm, UB/B 37.4%, 3.1 whorls. Site 18-6C, shell: H 0.74 mm, B 1.60 mm, H/B 0.46; AH 0.56 mm, AB 0.54 mm, AH/AB 1.04; UB 0.53 mm, UB/B 32.9%, 2.9 whorls. Although the 18-6C shell is smaller than the one from 18-2C, it has a fully formed peristome and may represent an adult; it is a fresh shell with intact peristome. The 18-2C shell has much of the periostracum missing from the dorsal surface and most of the aperture broken away. Shell discoid, with protoconch and spire scarcely raised above the top edge of the body- whorl. Whorls expanding slowly at first, with body whorl increasing more rapidly; suture deep. Body-whorl with distinct shoulder above the suture but otherwise with evenly rounded periphery. Umbilicus wide, open, almost symmetrical, internally exposing much of underside of protoconch and some of inner part of all whorls. Aperture broadly oval except where interrupted by penultimate whorl. Peristome thin, mainly plane but slightly reflected over extreme edge of umbilicus, its upper edge arching forwards beyond the outer and lower edges, its ventral end extending further around whorl than the distal end. Protoconch of 1.3- 1.5 whorls minutely granular or pitted or appearing smooth; teleoconch from about whorl 1.6 onwards with strong rather regular radial ribs (body whorl having ca 64 ribs on smaller shell, ca 57 on larger shell); each rib raised as a cord of even width that arcs backwards towards shell periphery, with its height greatest on periphery; periostracum between ribs nearly smooth but with fine grid-like pattern of 4-6 neat radial lines between each pair of ribs and running parallel to them, the radial lines being intersected by spiral lines. Sculpture on underside of teleoconch generally as on upperside, but radial lines between ribs tend to be stronger near aperture; similar sculpture also visible inside umbilicus on inside of whorls of spire. Protoconch translucent, becoming whitish on older shell, teleoconch light brown, slightly glossy; inside of aperture brown, translucent because shell thin. Known only from shells, so features of the exterior of the body, genital anatomy, etc., remain unknown. T. iredalei was described from Kenya (PRESTON, 1912); VERDCOURT (1991: 373 figs 12a-c) figured a lectotype. VAN BRUGGEN & VAN GOETHEM (2001: 161 figs 19-21) reported its range as D.R. Congo (now Zaïre), Kenya and perhaps Cameroon (the last based on two shells tentatively reported by DE WINTER & GITTENBERGER, 1998) and WRONSKI ET AL. (2014: 163, 165: table 1) reported it from Bioko. Our specimens from São Tomé thus represent the first confirmed records from oceanic islands off West Africa.

TRUNCATELLINIDAE Steenberg, 1925 Truncatellina R.T. Lowe, 1852

27

Truncatellina thomensis sp. nov. D. Holyoak & G. Holyoak (Fig. 2E)

Type material: Holotype NHMUK reg.no. xxxxxx, shell (H 2.26 mm, B 0.88 mm), 24 Nov. 2018, GAH, DTH, et. al., site 18-2C. Type locality: São Tomé Island, Morro Claudina (near Bom Sucesso), N0.2900° E6.6055°, 1257 m alt., on ground in montane rainforest (sieved from shelly debris at thrush anvil). Paratypes: From type-locality, collected with holotype, site 18-2C, 7 sh + 2 fragments; 27 Nov. 2018, Morro Claudina, N0.28909° E6.60442°, ca 1290 m alt., floor of montane forest with understorey (sieved from shelly debris at thrush anvil), GAH & DTH 18-6A, 3 sh; ditto, N0.29036° E6.60537°, 1254 m alt., floor of rather open montane forest with some alien trees (sieved from shelly debris at thrush anvil), GAH & DTH 18-6D, 9 sh. Etymology: The species epiphet thomensis is an adjective derived from the type-locality on São Tomé (formerly sometimes known as île de Saint Thomas). Description: H 1.71-2.26 mm, B 0.82-0.93 mm, H/B 2.04-2.58; body-whorl height 0.82- 1.05 mm (44.0-55.1% of H); AH 0.47-0.56 mm, AB 0.39-0.54 mm, AH/AB 1.0-1.27; whorls 5.7-6.1. Shell dextral, sub-cylindrical with sub-conical upper-spire (body-whorl and penultimate-whorl similar in width but peristome extends slightly further; whorl 4 slightly narrower; whorls 0-3 sub-conical with bluntly rounded apex); whorl height increasing ± gradually. Whorls somewhat rounded, but with weak shoulder below suture. Umbilicus a shallow chink (slit) under columellar side of peristome; back of body-whorl mainly rounded, but with short keel above umbilicus. Aperture broadly ovate to ovate-oblong except where interrupted by penultimate-whorl, the columellar border, palatal border, or both, often straighter than basal margin. Peristome thin to slightly thickened throughout, gently reflected on palatal margin, more strongly reflected on columellar margin, but sharply turned inwards to meet penultimate-whorl at top of palatal margin; outer palatal edge convex and rounded. In side view the peristome usually arched upwards along middle part of palatal margin. Aperture without any trace of teeth or denticles, even when viewed obliquely and none visible by translucence on concealed parts of body-whorl. Protoconch (whorl 1) smooth; teleoconch with numerous closely-spaced fine riblets or striae, which somewhat irregular, nearly straight, passing somewhat obliquely downwards to right (with shell viewed with spire pointed upwards and aperture on top). Protoconch whitish; teleoconch light-brown, glossy; interior of aperture of freshest shells light brown with the thin shell translucent, although columellar edge whitish; on older shells interior of aperture more extensively whitish, less translucent. Features of the exterior of the body, genital anatomy, etc., remain unknown. Discussion: The only previous reports of Truncatellina from islands in the Gulf of Guinea were by VAN BRUGGEN (1994: 23-24, figs 24, 25, 29, see pp. 377-378 for corrected version of figs) for T. adami van Bruggen, 1994 from Fernando Poo (now Bioko), but that species has a wider shell (B 1.0-1.1 mm) which is almost smooth on later whorls ("no sculpture beyond some very faint growth lines"). Elsewhere in West Africa, the genus was reported only by DE WINTER (1990: 306-307), who named T. silvicola as a new species, apparently still known only from its holotype, from 600 m altitude in forest of the western Ivory Coast. Compared to T. thomensis sp. nov., that species is smaller (H 1.64 mm), has the whorls more swollen, and ribs more widely spaced on the teleoconch. All other Truncatellina described from tropical Africa are known only from localities that are much more distant geographically from São Tomé: T. pygmaeorum (Pilsbry & Cockerell, 1933) from central Kenya through east and south-east Zaïre to south-west Angola and north- eastern South Africa has a sinistral shell (e.g. VAN BRUGGEN, 1994; HERBERT & KILBURN, 2004: 113) whereas T. thomensis sp. nov. has a dextral shell. T. naivashaensis (Preston, 1911) from Kenya has a fold on the columella visible when the shell aperture is viewed

28

obliquely (ADAM, 1954; VAN BRUGGEN, 1994); T. ruwenzoriensis Adam, 1957 has strong radial sculpture on the upper part of the spire and weak irregular sculpture on the body-whorl (ADAM, 1957: 6, fig. 5; VAN BRUGGEN, 1994); T. ninagongonis (Pilsbry, 1935) from Zaïre (L. Kivu in former Belgian Congo) and Malawi has fine striae on the shell that are stronger behind the peristome (ADAM, 1957: 3-4, 6 fig. 4; VAN BRUGGEN, 1994). T. upembae Adam, 1954 was figured and described by VAN BRUGGEN (1994: 19, figs 9-12, 22-23, 29, see pp. 377-378 for corrected version of figs) and reported to be widely distributed in Central Africa (NE. Angola, S., E. and NE. Zaïre) in that paper and by VAN BRUGGEN & VAN GOETHEM (2001: 154). Although different populations of T. upembae are rather variable in shell characters, and the overall shell measurements for that species broadly overlap those of T. thomensis, comparisons with the published figures shows that T. thomensis differs in having much closer and finer axial ribbing on the later whorls, the outer palatal edge of the peristome convex and rounded rather than slightly concave or straight, and usually a shallower suture.

VALLONIIDAE Morse, 1864 syn. Pupisomatidae Iredale, 1940 Stoliczka, 1873

Pupisoma dioscoricola (C.B. Adams, 1845)

Helix dioscoricola C.B. Adams, 1845, Proc. Boston Soc. nat. Hist., 2, p. 16: TL Jamaica. syn. Helix orcula Benson, 1850, Ann. Mag. nat. Hist., (2) 6, p. 251; TL India. New to São Tomé: 26 Nov. 2018, near Ponta Furada, N0.23665° E6.46775°, 240 m alt., flushed steep rock of trackside cutting, part-shaded by secondary forest, GAH & DTH 18-5, 1 sh; 27 Nov. 2018, Morro Claudina (near Bom Sucesso), N0.28909° E6.60442°, ca 1290 m alt., floor of montane forest with understorey (sieved from shelly debris at thrush anvil), GAH & DTH 18-6A, 1 sh; 28 Nov. 2018, Cascata de São Nicolau, N0.2857° E6.6257°, 885 m alt., around waterfall on steep rocky slopes with secondary forest, GAH, DTH et al. 18-7, 1 sh. Helix orcula is treated as a synonym of H. dioscoricola here following www.mollusca.base.org (accessed 7 July 2019). HERBERT & KILBURN (2004: 114-115) gave the range of Pupisoma orcula as extending widely over eastern South Africa, northwards to Zimbabwe, Mozambique and central Africa, and occurring elsewhere in , Hawaii and Australia. VERMEULEN & WHITTEN (1998: 83) placed it in the genus Parazoogenetes Habe, 1956 (an arrangement not accepted on MolluscaBase in 2019) and described the species as "common and widespread in Africa, Asia, Australia, Pacific". It was tentatively reported from West Africa by DE WINTER (1990: 307-308), who figured Pupisoma cf. orcula from the western Ivory Coast, based on a single shell from undisturbed forest habitat at 600 m alt. HAUSDORF (2007) revised New World Pupisoma, giving the range there of P. dioscoricola as extending from southern U.S.A. through the Caribbean and Central America to the Galapagos Is., S. Brazil and N. Argentina.

Pupisoma harpula (Reinhardt, 1886)

Helix (Acanthinula) harpula Reinhardt, 1886, Sitzungsb. Ges. Naturf. Fr. Berlin, p. 115; TL Tokio, Kanda [Japan]; PILSBRY (1926, Man. Conch., 2 (27) p. 198) recombined it as Zoögenetes (?) harpula (Reinh.) but this was corrected by HAAS (1937: 10-11, pl. 3 figs 25- 27) who figured "Typus SMF 3443". Syn. Pupisoma japonicum Pilsbry, 1902, Nautilus, 16, p. 21; TL Hirado, Hizen [Japan].

29

New to São Tomé: 24 Nov. 2018, Morro Claudina (near Bom Sucesso), N0.2900° E6.6055°, 1257 m alt., montane rainforest (sieved from shelly debris at thrush anvil), GAH, DTH et al. 18-2C, 5 sh; 27 Nov. 2018, Morro Claudina, N0.28909° E6.60442°, ca 1290 m alt., montane forest with understorey (sieved from shelly debris at thrush anvil), GAH & DTH 18-6A, 1 sh; ditto, N0.2890° E6.6044°, sieved from ground-litter on floor of montane forest, sample 18-6B, 1 sh; ditto, N0.29036° E6.60537°, 1254 m alt., floor of rather open montane forest with some alien trees (sieved from shelly debris at thrush anvil), sample 18- 6D, 1 sh; 28 Nov. 2018, Cascata de São Nicolau, N0.2857° E6.6257°, 885 m alt., around waterfall on steep rocky slopes with secondary forest, GAH, DTH et al. 18-7, 3 sh; 9 Dec. 2018, path from above Bom Sucesso Botanical Garden towards Lagoa Amélia, ca N0.2823° E6.5964°, 1290-1415 m alt., montane forest (sieved), GAH & DTH 18-21, 1 sh. PILSBRY (1920: 25-26, pl. 2, figs 11, 12) again treated this species as P. japonicum, separating it from P. orcula by the wider umbilicus (one-seventh of shell diameter) and giving its range as Japan and South Africa. HERBERT & KILBURN (2004: 114-115) described and figured both P. harpula and P. orcula (regarding P. japonicum as a synonym of P. harpula). They noted that they are very similar, but that P. harpula differs in being smaller than P. orcula (up to 1.65 mm, cf. 2.0, rarely 2.2 mm), with shell more elevated, having a wider umbilicus and lacking spiral sculpture. They gave the range of P. harpula as extending from Japan to central and southern Africa, reaching the southern Cape, occurring in similar habitats in S. Africa to P. orcula and sometimes living with it on the same tree. VERDCOURT (2006) listed P. harpula for East Africa. This species has also been reported as P. japonicum from Central Africa (e.g. ADAM, 1954: 806), E. Madagascar (EMBERTON ET AL., 2010) and the Mascarene Is. (GRIFFITHS & FLORENS, 2006: 86).

AGRIOLIMACIDAE H.Wagner, 1935 Deroceras Rafinesque, 1820

Deroceras laeve (O.F. Müller, 1774)

Limax lævis O.F. Müller, 1774, Verm. terr. fluv., 2, p. 1; TL restricted by WIKTOR (2001: 107) to Frideriksdal near Copenhagen (Denmark). New to São Tomé: 3 Dec. 2013, on path between Bom Sucesso and Macambrará, near Rio Manuel Jorge, N0.2847° E6.6097°, ca 1149 m alt., trackside, cultivation edges, scrub & small forest patches, GAH & DTH ST3, 6 spm (2 dissected were both aphallic, but with spermoviduct well developed); Mar. 2017, surroundings of Jardim Botânico at Bom Sucesso, non-forested, with bananas & fields of vegetables, leg. Martina Panisi #1, 5 spm; ditto, #2, 6 spm; 24 Nov. 2018, near Quinta da Graça, N0.29921° E6.64174°, 678 m alt., scrub, cultivated gardens, roadside banks, few trees, GAH & DTH 18-3, 1 examined alive (not kept); 2 Dec. 2018, just S. of Milagrosa, on the road between Trindade and Bombaim, N0.27667° E6.65986°, 398 m alt., tall secondary forest on slopes, track edges with flushed rocky cuttings & abandoned cocoa plantations, GAH & DTH 18-10, examined alive (not kept); 10 Dec. 2018, in Bom Sucesso Botanic Garden, N0.2886° E6.6121°, 1156 m alt., botanic garden with groves of trees, lawns & cultivated patches, GAH & DTH 18-22, examined alive (not kept); 10 Dec. 2018, on path between Bom Sucesso and Macambrará, near Rio Manuel Jorge, N0.28571° E6.60912°, 1114 m alt., horticulture areas with bananas, groves of scrub & trees alongside track, GAH 18-23, examined alive (not kept). WIKTOR (1999: 463) noted that "the original distribution range of D. laeve seems to be the Palearctic, which is practically wholly inhabited by the species, from the subpolar zone up to its southern fringes, including China. Owing to its short life cycle and high ecological tolerance, the slug is extraordinarily susceptible to introduction. It can be encountered on all

30

continents, except for Antarctic. It occurs quite locally but all over the globe, even on tropical islands such as New Guinea or others in the Pacific".

EUCONULIDAE H.B. Baker, 1928

Afroconulus Van Mol & van Bruggen, 1971

Afroconulus roseus sp. nov. D. Holyoak & G. Holyoak (Figs 2C, 7)

Type material: Known only from the unique holotype, NHMUK reg.no, collected 10 Dec. 2018, by DTH (with FS & MP), site 18-24. Type locality: São Tomé Island, path from above Bom Sucesso Botanical Garden to Lagoa Amélia, N0.2835° E6.5981°, ca 1353 m alt., found alive ca 1.5 m above ground on large glossy green leaf of herb in understorey of montane rainforest. Etymology: The genus Afroconulus has been treated as a masculine noun. The epiphet roseus is from the Latin (meaning pink), alluding to pink dorsal coloration of the body of the living snail.

Description: Shell B 4.44 mm, H 4.06 mm, H/B 0.91; AB 2.35 mm, AH 2.18 mm, AH/AB 0.93; whorls 5.0. Dextral, conical, convex below; protoconch small, whorls increasing regularly; suture moderately deep. Body-whorl rounded, with definite angle just above periphery that becomes less obvious near aperture. Aperture broadly oval, except where interrupted by penultimate whorl. Peristome thin, plane except short columellar part which is reflected to conceal the tiny umbilicus. Protoconch with microsculpture of fine close radial riblets intersected by spiral ridges. Teleoconch above and below periphery with fine closely-spaced spiral ridges, intersecting coarser, less regular, ill-defined, low riblets and lines. Shell very thin, fragile, translucent; pale brown; spire appearing to have waxy lustre that becomes glossy on body whorl, especially the underside. Exterior of body described from living animal as having striking pattern produced by blackish ommatophores and their retractor muscles, blackish spot dorsally at tail tip and bright pink dorsal surface of other exposed parts of body, shading to lighter pinkish on face and whitish on flanks, foot-fringe and sole of foot. The mantle visible through the translucent shell was mainly whitish but marked with a strong blackish line along the mantle-collar. Body of holotype drowned and preserved in ims (partly contracted) was 7.1 mm long, of which tail formed 3.7 mm. All trace of pink coloration has vanished. Sole of foot narrow, tripartite, white; front of foot separated from base of head by deep transverse groove (as in other Helicarionoidea studied). Thin groove bordering upper edge of foot-fringe continuous along sides of body and around tail. Tail appears long, rather narrow (higher than wide), white with light grey dorsal area around tip; no distinct caudal horn, only a slight subapical protuberance; caudal fossa visible only as very small pore high on steep end of tail. Front of head with two pairs of tentacles, the lower pair short. Mantle-collar raised around front edge and on sides of body as thin continuous flap, narrowing posteriorly on both sides and absent above base of tail; lacking separate mantle-laps; white, except for blackish line along the front right-hand half. Mantle inside shell white with two black blotches towards rear of dorsum (Fig. 7D). Genital pore tiny, on front right-hand flank of body at one-third height, a short distance behind base of upper tentacle. Genital atrium a short but rather wide tube, dividing proximally into distal end of penis and distal end of vagina. Penis a short tube only twice as long as wide, with opaque wall. Distal epiphallus an opaque narrow white tube enclosed in a translucent thin muscular cylindrical sheath almost twice as wide that appears well defined at

31

both proximal and distal ends. Proximal part of epiphallus tightly convoluted in situ, with strands of connective tissue joining its folds; comprised of three short wide portions separated by two narrower necks; penial retractor muscle widely attached near proximal end of epiphallus, extending onto inside of mantle wall; proximal end of epiphallus narrows into vas deferens; flagellum lacking. Vas deferens starts near distal end of spermoviduct, passes distally alongside free oviduct and then returns proximally to reach only the proximal part of epiphallus (not passing into vaginal-penial angle). Vagina just over twice as long as wide, somewhat swollen, with distal end narrower; free oviduct at least three times length of vagina, stout, the distal two-thirds thicker-walled because encased in an opaque cylindrical gland. Duct of bursa copulatrix a tube somewhat narrrower than free oviduct or epiphallus- sheath, lying alongside free oviduct; reservoir of bursa not distinct from the duct, its proximal apex reaching distal end of spermoviduct, with a thread-like strand extending proximally from tip of reservoir to inner edge of mid-part of spermoviduct. The spermoviduct large and stout with distinct folds, attaching proximally to large curved albumen-gland that tapers to bluntly pointed apex. Part of broken-off common hermaphrodite duct remains on inner surface of albumen gland towards its inner, distal end. Right ommatophore retractor muscle passing through distal genitalia between epiphallus-sheath and duct of bursa copulatrix. Comments: The genus Afroconulus was named by VAN MOL & VAN BRUGGEN (1971: 286) with Sitala diaphana Connolly, 1922 as type species by original designation. DE WINTER & VAN BRUGGEN (1992) recognised three genera of Euconulidae in Africa (Afropunctum F. Haas, 1934, Afroguppya de Winter & van Bruggen, 1992 and Afroconulus). Among these genera, Afroconulus is distinctive anatomically in possessing a well developed vagina; in having most of the free-oviduct surrounded by a perivaginal gland; in having most of the length of the penial epiphallus enclosed in a long cylindrical muscular sheath, and in lacking a penial flagellum (VAN MOL & VAN BRUGGEN, 1971; DE WINTER & VAN BRUGGEN, 1992; SCHILEYKO, 2002a: 1135-1136). Whereas both Afropunctum and Afroguppya are known to bear live young because shelled embryos are found in the free oviduct (DE WINTER & VAN BRUGGEN, 1992), Afroconulus may differ in laying eggs, since HERBERT & KILBURN (2004: 255) refer to "tiny, round, white eggs" as "retained within the parent for some time" in A. diaphanus; although the distal female tracts of the mature holotype of A. roseus were not opened they certainly lacked shelled embryos. Helix jenynsi L. Pfeiffer, 1846 has also been placed in Afroconulus, but it is now regarded as Sitala jenynsi (L. Pfeiffer, 1846) and placed in the Helicarionidae on MolluscaBase (http://www.marinespecies.org/aphia.php?p=taxdetails&id=1262369, accessed 13 Aug. 2019). The shell characters and distal genital anatomy of A. roseus sp. nov. generally match those of Afroconulus diaphanus (Connolly, 1922) (VAN MOL & VAN BRUGGEN, 1971; SCHILEYKO, 2002a: 1135-1136; HERBERT & KILBURN, 2004: 255). However, it is treated as a distinct species here because A. roseus has the shell lower than wide (H/B 0.91 in holotype), whereas figures of A. diaphanus show its shell is higher than wide (H/B 1.05), Also, comparison with the figure by SCHILEYKO (2002a: 1136 fig. 1494B) shows that the proximal part of the epiphallus (part outside the sheath) of A. roseus is much longer compared to the sheath length than in A. diaphanus (>0.5× sheath, cf. <0.3×); its proximal epiphallus has three wide portions separated by two narrow "necks" (Fig. 7B, C) and is tightly convoluted in situ, not a single wide portion which is C-shaped; also, the reservoir of the bursa copulatrix is not swollen so as to be distinct from adjoining parts of the bursa duct. All these characters of course need to be compared from more specimens than the unique holotype of A. roseus, but the large gap in geographical distribution between its type locality in montane rainforest on São Tomé and the range of A. diaphanus (montane forests from Kenya to Transkei: HERBERT

32

& KILBURN, 2004: 255) also encourages us to regard it at least provisionally as a distinct species.

Afropunctum F. Haas, 1934

Afropunctum seminium (Morelet, 1873)

Helix seminium Morelet, 1873, J. Conchyl., 21 (4), p. 329; TL "prov. Gabon". BREURE ET AL. (2018: 424, fig. 1015) figured a syntype from NHMUK. New to São Tomé: 7 Dec. 2013, path to Lagoa Amélia from Bom Sucesso Botanical Garden, N0.2825° E6.5969°, ca 1400 m alt., montane forest on ridge, with closed canopy & understorey of saplings, GAH & DTH ST8, 2 sh; 13 Dec. 2013, by EN2 just NE of Monte Mário "roça", N0.0606° E6.5561°, ca 236 m alt., forest on slopes by road with understorey of saplings & ferns beneath, GAH & DTH ST13, 1 sh; 27 Nov. 2018, Morro Claudina (near Bom Sucesso), N0.28909° E6.60442°, ca 1290 m alt., floor of montane forest with understorey (sieved from shelly debris at thrush anvil), GAH & DTH 18-6A, 3 sh; ditto, N0.29181° E6.60592°, 1288 m alt., floor of montane forest near tiny clearing (sieved from shelly debris at thrush anvil among buttress roots of tree), sample 18-6C, 3 sh; ditto, N0.29036° E6.60537°, 1254 m alt., floor of rather open montane forest with some alien trees (sieved from shelly debris at thrush anvil), sample 18-6D, 1 sh; 28 Nov. 2018, Cascata de São Nicolau, N0.2857° E6.6257°, 885 m alt., around waterfall on steep rocky slopes with secondary forest, GAH, DTH et al. 18-7, 6 sh & 1 spm; 1 Dec. 2018, just before the Cascata de Bombaim (Rio Abade), on the road between Trindade and Bombaim, N0.2534° E6.6318°, 485 m alt., flushed rock of low trackside cutting, part-shaded by bamboo & secondary forest, GAH, DTH & RFL 18-9, 1 spm; 2 Dec. 2018, just S of Milagrosa, on the road between Trindade and Bombaim, N0.2766° E6.6598°, 398 m alt., secondary forest on slopes, abandoned cocoa plantations, trackside rock cuttings, GAH & DTH 18-10, 44 sh & 7 spm; 9 Dec. 2018, path from above Bom Sucesso Botanical Garden towards Lagoa Amélia, ca N0.2823° E6.5964°, 1290-1415 m alt., montane forest, GAH & DTH 18-21, 4 sh. New to Príncipe: 5 Dec. 2018, 1.5 km SW of Terreiro Velho, close to Oquê Pipi, N1.59965° E7.41082°, 308 m, tall native forest, GAH 18-16A, 1 sh. The identification characters, systematic position and geographical range of this species were reviewed by DE WINTER & VAN BRUGGEN (1992), who also provided good illustrations of its shells, genital anatomy and radula. Their study revealed that the species has an extensive range in tropical Africa, with west African records from countries including Ghana, Nigeria and Gabon, but no reports hitherto from offshore islands.

HELICARIONIDAE Bourguignat, 1877

The supra-generic classification by BOUCHET ET AL. (2017) recognised three families within the Helicarionoidea Bourguignat, 1877 as Helicarionidae, Godwin-Austen, 1888 and Urocyclidae Simroth, 1889. Although the slug genus Dendrolimax Heynemann, 1868 clearly belongs in Urocyclidae subfamily Urocyclinae and the large thick-shelled snails of the genus Rhysotina endemic on São Tomé form the subfamily Rhysotininae Schileyko, 2002 of the Urocyclidae, allocation of the remaining mainly endemic snails and semi-slugs from São Tomé and Príncipe to families is less clear. This is partly because the genital anatomy of the four hitherto known species has remained undescribed except for a single small drawing for Nanina thomensis Dohrn in a largely overlooked paper (GIRARD, 1895: 32-33) otherwise devoted to Thyrophorella thomensis Greeff, 1882. The present study therefore describes the genital anatomy of three of

33

the four known species along with that of four new species. This reveals that they form five undescribed and presumably mainly endemic genera each with distinctive genital anatomy. It is difficult to place some of them within the recent classifications by VAN MOL (1970), VAN GOETHEM (1977), SCHILEYKO (2002b) and DE WINTER (2008). The present arrangement is therefore provisional and based on overall resemblance of characters of the distal genitalia, since it is clear that the helicarionoid group is in need of a phylogenetic analysis using both morphological and molecular data (DE WINTER, 2008: 474). For Helicarionoidea, terminology of the mantle lobes and other external soft parts used here follows SOLEM (1966) and DE WINTER (2008). In all taxa described below the tail was longer than the forepart of the body; the foot sole was tripartite with a narrow central zone; the front edge of the sole separated from base of head by deep transverse groove. All these taxa also had two long upper tentacles with a small black eye-spot near the tip when tentacle is expanded and two short hemispherical to very shortly cylindrical lower tentacles. Head warts or other protuberances on the front of the head were not seen.

Apothapsia gen. nov. D. Holyoak & G. Holyoak

Type species: Nanina thomensis Dohrn, 1866 (see below). Etymology: Based on snail genus Albers, 1860 of Helicarionidae, originally published as a subgenus of Nanina Gray, 1834 non Risso, 1826. Albers (in MARTENS, 1866: 56) did not state the derivation of the word Thapsia, but it appears to be Greek in origin. The new generic name therefore combines the Greek Apo- (meaning apart, or separate from) with Thapsia. Hence, like Thapsia, it is a feminine noun. Material examined: >100 specimens of A. thomensis, from 16 localities on São Tomé, all in CGAH. Description: based only on type species:

Apothapsia thomensis (Dohrn, 1866) (Figs 9A, B, 10H-J)

Nanina thomensis Dohrn, 1866 (DOHRN, 1866a: 114, pl. 5 figs 8-10), treated as Thapsia thomensis Dohrn in later literature (e.g. GERMAIN, 1908: 60; 1916: 216-218, pl. 10 figs 5-7).

Description: Shell dextral, depressed-conical to nearly conical, with rounded spire and flatter base. At maturity, H 7.5-11.0 mm, B 10.4-13.0 mm, with 4.2-4.8 whorls. Protoconch small; whorls of spire rounded with shallow suture (which forms conspicuous narrow band or double line), increasing gradually, but with relatively larger body whorl that is evenly rounded in profile or with slight but definite angle forming upper edge of its periphery. Aperture rounded-oval except where interrupted by penultimate whorl; peristome thin, plane except for narrow reflected area around about half of umbilicus. Umbilicus very narrow (< 1 mm) but deep and usually exposing interior of spire. Shell thin, rather fragile, translucent when fresh but becoming opaque and whitish when old. Fresh shells pale brown to yellow- brown above, paler to whitish beneath, very glossy throughout. Protoconch smooth; teleoconch with inconspicuous sculpture comprised of faint irregular radial lines (mainly close below suture) and very faint irregular spiral microsculpture. Body relatively long and slender, long tail ending dorsally in short-conical caudal horn (ca 1.5 mm long, raised, or near-horizontal and directed backwards in live snails), with large fossa in rear end beneath. Mantle-collar wide, with left shell-lap and right shell-lap both large, thin, together covering much of exterior of shell of undisturbed live snails including top of spire, hence leaving minority of shell surface visible as broad arc (Fig. 10H); these shell-laps are withdrawn inside shell when snails are approached closely or touched. Snails

34

preserved in ims have exposed parts of body mainly whitish to pale pink, top of head and upper foreparts pale grey in some individuals; shell-laps dull pale pinkish to pale pinkish- grey; mantle-collar and mantle surface inside body-whorl and penultimate whorl marked with fine speckling and lines giving grey ground-colour (blackish-grey in darkest individuals), with large spots or conspicuous blotches of white. When living snails are disturbed the body flushes with orange-pink (Fig. 10J); when they are handled copious orange body-slime is produced that briefly stains human hands and the water in which specimens are drowned. Distal genitalia were studied in six individuals, all from different localities. Genital pore on right-hand side of forepart of body, behind lower tentacle, directly beneath base of upper tentacle. Genital atrium short, dividing proximally on one side into distal end of penis, on opposite side into distal end of vagina, with narrow but well-defined short band of muscle attached to its outer proximal wall between those organs. Penis large, appearing broadly cylindrical in outline within a thin flexible outer sheath covering all but most distal part; coiled inner contents clearly visible through central portion of sheath. In two fully mature snails the distal penis and penis sheath were opened longitudinally, both specimens revealing (Fig. 9A-C): a short wide penis in distal part with thick muscular walls; this passing into central portion with narrower epiphallus coiled in situ; this in turn continuing into more slender epiphallus in proximal one-third of penis complex, where its tight and rather irregularly intertwined coils were firmly bound together with threads, bands and small sheets of strong connective tissue suggestive of spider's web; at proximal end a small epiphallic caecum with penis retractor muscle attached terminally (muscle broken off in one snail; long, slender and passing proximally to columellar musculature in the other). Flagellum lacking. Vas deferens a slender tube throughout, passing from near distal end of spermoviduct distally alongside free oviduct then through vaginal-penial angle, returning proximally alongside exterior of penis sheath to pass through it close before proximal end of penis outer sheath, and so continue into proximal end of epiphallus. Vagina cylindrical, about two-thirds length of free oviduct, somewhat convoluted, with strong muscular wall; free oviduct of similar thickness to vagina, with thick muscular wall. Duct of bursa copulatrix short, less than half length of vagina and thinner than vagina; reservoir of bursa shortly ovoid, positioned alongside free oviduct, with thin strand of muscle arising at its proximal tip and passing proximally to join central part of prostate which is adherent to columellar side of spermoviduct. Right ommatophore retractor muscle passes through angle between distal penis and distal vagina. Comments: The study by DE WINTER (2008) began the process of revising the large and heterogeneous assemblage of species that have been placed in Thapsia Albers, 1860 mainly on the basis of genital anatomy and to a lesser extent on shell characters. He did not discuss A. thomensis; although the figure of its genital anatomy by GIRARD (1895: 32-33) was accurate, it did not portray any features inside the penis sheath. Comparison of our observations with those of DE WINTER (2008) shows that A. thomensis resembles Thapsia s.s., Saphtia de Winter, 2008 and Pseudosaphtia de Winter, 2008 in having a distinct vagina and long penial epiphallus, with penial retractor muscle arising from end of a short epiphallic caecum. However, A. thomensis differs from all taxa placed in those genera in having a much smaller bursa copulatrix on a much shorter duct (so the reservoir lies alongside free oviduct not central part of spermoviduct), not having a penial flagellum, and especially in having the very long proximal part of the epiphallus tightly bound into an ovoid mass inside the penis sheath, rather than free or much more loosely convolute in situ. Despite having a very similar shell, Vanmolia de Winter, 2008 differs markedly from all of these genera in lacking a vagina (so duct of bursa copulatrix and free oviduct both arise from genital atrium), penis with a

35

large verge internally and a shorter epiphallus, characters perhaps suggestive of Urocyclidae rather than Helicarionidae. In A. thomensis the total length of penis + epiphallus in situ before dissection was ca 12 mm, but after complete dissection of the very long epiphallus of the figured individual it was ca 37 mm (>3 × shell diameter; ca 2 × total length of extended body). However, only the distal part of the penis is likely to be everted during mating, so that the length of the intromittent organ will be 5-6 mm, approximately corresponding to the length of the vagina. The very short duct of the bursa copulatrix might imply that the (unknown) spermatophore is equally short, but this needs further study since a long epiphallus would normally imply that the spermatophore is equally long. The following sp. nov. is unknown anatomically and only tentatively referred to Apothapsia. Hence, only the type species Apothapsia thomensis, is securely referred to this genus. It is an endemic on São Tomé, where it is probably the most common in wooded habitats on the island, occurring in montane primary-forest, lowland secondary- forest and plantations, from near sea-level to at least 1400 m alt. Living snails are faster- moving than most land snails and climb freely on foliage of herbs, ferns, saplings, or banana plants, and on dead wood and tree trunks. They also rest under dead wood lying on the ground.

Apothapsia moreleti (Germain, 1915)

Trochonanina (Trochozonites) Moreleti Germain, 1915 (GERMAIN, 1915: 287; 1916: 221-223, pl. 9 figs 6-8, 12). Trochonanina (Trochozonites) Moreleti var. fasciata Germain, 1915 (GERMAIN, 1915: 287; 1916: 222).

Material examined: All from São Tomé Island: 12 Dec. 2013, by EN1 between Lagoa Azul [Morro Carregado is the name on the military maps] and Praia Mutamba, N0.4006° E6.6075°, ca 5 m alt., dead/drifted shells from soil surface & debris by small dry stream- course above shore in lower edge of valley with cover of dry woodland, GAH & DTH ST12, 1 sh & 1 sh fragment; 24 Nov. 2018, near Quinta da Graça, N0.29928° 6.64174°, 678 m alt., scrub, gardens, roadside banks, with few trees, GAH & DTH 18-3, 1 sh; 2 Dec. 2018, São Tomé Island, just S of Milagrosa on the road between Trindade and Bombaim, N0.2766° E6.6598°, 398 m alt., secondary forest on slopes, abandoned cocoa plantations, trackside rock cuttings. GAH & DTH site 18-10, 1 sh; 11 Dec. 2018, by S. side of EN2 between the “roças” Fraternidade and Soledade (ca 1.6 km due S. of São João dos Angolares, centre), N0.1262° 6.6371°, 22 m alt., old cocoa plantations & bananas, under tall trees & palms, near stream (Água São Pedro), GAH & DTH 18-26, 1 sh.

Description: Based on four shells (+ one fragment) from four localities. H 4.02-4.42 mm, B 9.31-10.32 mm, H/B 0.42-0.45; AH 3.28-3.75 mm, AB 4.49-5.09 mm, AH/AB 0.73-0.76; UB 0.40-0.47 mm (UB/B 4.3-5.0%); whorls 4.2-4.6. Shell dextral, very depressed-conical with flatter base and rounded spire. Protoconch small, whorls of spire increasing regularly in size, slightly rounded above; suture shallow and appearing to form distinct narrow band or double line. Body-whorl not much wider than last part of penultimate whorl, with prominent angle at upper edge of periphery forming distinct keel; top edge of keel concave, defining a rather indistinct narrow peripheral cord. Aperture oval in outline, with angle formed at periphery by keel and parietal area widely interrupted by penultimate whorl. Peristome thin, plane throughout except for narrow reflection of columellar margin near umbilicus. Umbilicus narrow, deep, exposing interior of spire, symmetrical, slightly overhung by edge

36

of peristome. Protoconch somewhat corroded but remaining periostracum smooth; teleoconch with inconspicuous sculpture of fine irregular radial lines and very fine irregular microsculpture of spiral lines; periostracum of body-whorl also with minute closely-spaced granular or decussate pattern just visible at ×56 under intense light from low angles. Shell thin, apparently fragile, translucent when fresh. Protoconch whitish, teleoconch light brown and glossy throughout. Because only empty shells have been found, features of the exterior of the body, genitalia, etc. remain unknown. Discussion: The taxonomic position of this species in Apothapsia is provisional, awaiting revision when the genital anatomy can be described, since distinct genera that appear to be allied to Thapsia show few differences in shell characters (DE WINTER, 2008). Shells of A. moreleti differ from those of A. thomensis in having an obvious peripheral keel, whereas the keel in the latter species is commonly absent and at most very weak. The other consistent difference from A. thomensis is the darker brown shell coloration, while the flatter shells with lower spire, less swollen whorls of the spire and smaller adult shell size are almost consistent. All four of our localities for A. moreleti are in lowland habitats (5-678 m alt.), with altered vegetation, with no records from intact native forests, whereas A. thomensis occurs up to at least 1400 m and it is plentiful in montane primary-forest as well as in lowland plantations. A. moreleti also appears to be much scarcer than A. thomensis, with single shells from three localities and two shells from the fourth. A. thomensis was present alongside T. trindadensis at three of the four localities for the latter species (but unrecorded at 18-10, the type-locality). The disturbed habitats of A. moreleti probably increase the possibility that this taxon is an introduction to São Tomé from elsewhere in tropical Africa, but no obvious match for it has been found, its few sites were geographically well separated on the island and native endemic species were present at the same localities. The two papers by GERMAIN (1915: 287; 1916: 223) gave partly conflicting localities on São Tomé for his new species, although all were in lowlands (700 m alt. at most). He also noted (GERMAIN, 1916: 223) that the species had probably been described but not named by MORELET (1868: 57). Var. fasciata Germain, 1915 was described as having "agrémenés, au dernier tour, d'une étroite fascie brune rigoreusement appliquée sur la carène et visible, par transparence, à l'intérieur de l'ouverture" and also noted as "beaucoup plus rare que le type". None of our specimens has a brown shell band, which would be unusual in any genus allied to Thapsia. There is a possibility therefore that he had based var. fasciata on misidentified immature shells of the taxon we describe below as Thomitrochoidea trindadensis, which are keeled and often possess a brown band such as he described, but we think this unlikely because that taxon differs in shell size, shape and sculpture.

UROCYCLIDAE Simroth, 1889, Subfamily Sheldoniinae Connolly, 1925

Principicochlea gen. nov. D. Holyoak & G. Holyoak

Type species: Principicochlea tenuitesta sp. nov. Etymology: Principi- is derived from the name of the island of Príncipe; combined with cochlea (snail shell, Latin, feminine). Like cochlea, the new name is thus a feminine noun.

Description: the new genus is based only on the following species:

Principicochlea tenuitesta sp. nov. D. Holyoak, G. Holyoak & F. Sinclair (Figs 3B, 8A-F, 10E-G)

37

Type material: Holotype NHMUK reg.no., shell (B 8.38 mm) + bod, collected 19 Apr. 2019 by FS, A. Pedronho and A. Andrade. Type locality: Príncipe Island, high on Pico Príncipe, N1.58202º E7.38262º, 860 m alt., in low mossy forest. Paratypes: 1 sh (damaged) + bod, 6 spm in ims and 1 spm in 96% ethanol, all collected with the holotype on same day at type locality. Etymology: The species epiphet derives from tenui- (meaning thin, Latin adjective) and testa (shell, Latin, feminine). The new species epiphet tenuitesta thus agrees with the feminine generic name with its ending unchanged.

Description: Shell of holotype: H 4.29 mm, B (max.) 8.38 mm, B (min.) 6.97 mm, H/B 0.51, AH 4.22 mm, AB 4.96 mm, AH/AB 0.85, with 3.1 whorls; damaged shell of a paratype: B (max.) 7.91 mm, B (min.) 6.37 mm, with 3.0 whorls. Shell dextral, subglobular, with low rounded spire and flatter base, oval in outline. Protoconch small, whorls increasing rapidly and regularly, so body-whorl forms most of shell; whorls rounded at periphery; suture shallow, appearing as narrow band or double line. Aperture rounded-oval except where interrupted by penultimate whorl, in ventral view aperture not extending beyond columellar axis, so interior of spire not visible in ventral view. Periostracum thin, plane except for minute reflected band close to columellar axis, the outer and lower palatal margins with narrow membranous fringe that shrinks when shell dries. Umbilicus absent. Sculpture low and weak on whole shell. Strong oblique lighting reveals protoconch with low closely- spaced radial riblets; teleoconch also with low irregular radial riblets and lines, with weak and very irregular spiral lines in few places (mainly near suture, visible at ×56). Protoconch whitish, teleoconch dull yellow, glossy throughout. Shell very thin, transparent, uncalcified, flexible when moist, fragile and somewhat deformed when dry. Body long relative to that of small shell, which is carried on body with the spire on right- hand side. Body of relaxed snails preserved in alcohol 18.5-20.0 mm long, of which tail forms up to 11 mm and slightly contracted forepart forms up to ca 9 mm. Tail somewhat higher than wide, rounded dorsally, tapering towards tip, the end of its dorsum with caudal horn forming low blunt papilla, beyond which extreme tip of tail appears rounded in dorsal view. Mantle lobes extend to cover a small part of exterior of right-hand side and front of shell when living snails are active and disturbed (Fig. 10E-G) and also in some specimens preserved in alcohol (Fig. 8B, E), but it is unclear whether they cover more of shell when the snails are undisturbed. Drowned and relaxed specimens in alcohol have four lobes that can be pulled outwards from the mantle edge (Fig. 8D), (1) a thick low frontal shell-lap, (2) low thick crescentic lobe 2.0 mm long on front right-hand side of body below #1, (3) a left shell- lap extending above #1 around front of body and continuing for ca 5 mm to end in point directed outwards and posteriorly on left-hand side of body, (4) a large thin oblong right shell-lap 5.5 mm long, above and mainly posterior to #2 on right-hand side of body. Sole of foot white. Exterior of body whitish, but light grey-brown dorsally on upper one-third to one- half of tail (the dark colouring widest posteriorly), top of foreparts and front of head pale brown, the ommatophore retractor muscles giving darker grey-brown streaks visible through translucent skin. Mantle surface inside shell dull yellow on outer part of body-whorl with few irregular blackish streaks, on inner part of body-whorl dark olive-brown with bold irregular bands of light yellow forming variable network of coarse markings. Digestive gland inside whorls 0-1.5 pale brown; a band of light yellow extending alongside inner suture to protoconch. Genital anatomy studied in holotype and a paratype. Genital pore located near mid-height on right-hand side of fore-part of body close behind base of upper tentacle. Genital atrium very short, dividing proximally into a long atrial appendage on one side, the distal penis on

38

the other side, with distal vagina inserting between these organs. Atrial appendage very large (ca 8.5 mm long; longer than penis + epiphallus, equalling vagina + free oviduct, wider than both distal male and distal female tracts), thick-walled, with slender retractor muscle attached to proximal end; one specimen with thickened collar evident near distal end of appendage, from which a cylindrical stimulator with bluntly conical tip was extruded. Penis a long stout tube with muscular wall inside a thin outer sheath; epiphallus rather thin-walled throughout, curved, distal portion wider than penis, proximal portion thinner and separated by constriction (with outer wall of proximal portion attached by connective tissue to outer wall along proximal end of penis); epiphallic caecum near middle of curved epiphallus with retractor muscle attached to end of it which extends to insert on inside wall of diaphragm. Flagellum lacking. Vas deferens a thin tube, starting near distal end of spermoviduct, passing distally alongside free oviduct and vagina to near distal end of penis, returning proximally alongside penis to enter end of epiphallus. Vagina moderately long (somewhat longer than penis),with two bends when in situ, longer and stouter than free oviduct. Spermoviduct rather short (about equal to penis length); albumen gland not seen. Gonad inside digestive-gland close to apex of spire, linked by convoluted tube of common hermaphrodite duct to near proximal end of spermoviduct. Duct of bursa copulatrix a stout cylinder just proximally to where it arises from junction of vagina with free oviduct, becoming thinner proximally. Bursa copulatrix shortly cylndrical, distinctly wider than adjacent portion of its duct, the reservoir appressed to side of distal half of spermoviduct. Right ommatophore retractor passes between distal penis and vagina. Comments: Vitrina dumeticola Dohrn, 1866 (DOHRN, 1866b: 119, pl. 5, figs 1-4) was named from Príncipe "im Norden der Insel auf der Pflanzung Azeitona". CROSSE (1888b: 299-300) noted: "Hab. Azeitona, plantation située au Nord de l´île du Prince (H. Dohrn). Obs. Cette espèce vit sur les feuilles d'arbres, à une hauteur de 4 à 10 pieds. L'animal est jaune avec des tàches blanchâtres: il rampe avec vivacité H. Dohrn)." Germain (1916: 161) treated it as Helicarion (Africarion) dumeticola Dohrn, doubtless based on the reasonable assumption that the shells represent a vitriniform semi-slug from the Helicarionoidea rather than a member of the . However, P. tenuitesta is clearly different to Dohrn's Vitrina dumeticola, which has apparently not been refound. DOHRN (1866b: 119, pl. 5, fig. 1) figured a shell with the aperture in ventral view extending beyond the columellar axis as a large rounded sinus, exposing the inside of the penultimate whorl of the spire, a feature lacking in the present species. Furthermore, his text confirmed this in detail as he noted "columellaris recedens, valde arcuatus; margo anfractuum internus conspicuus" and "Unter den mir bekannten Arten steht ihr V. nitida Gould ziemlich nahe, auch unsere V. diaphana Drap. ähnelt ihr in der form", i.e. with receding columellar margin resembling that in Madeirovitrina nitida (Gould, 1847) and (Draparnaud, 1805) as shown in his pl. 5 fig. 1. Principicochlea tenuitesta is placed in a new genus here because it shows a distinctive combination of characters of the genital anatomy different to those among the numerous other helicarionoid semislugs known from tropical Africa (cf. VAN MOL, 1970; SCHILEYKO, 2002b). Thus, no other African genus combines: penis with well-developed epiphallus and retractor muscle arising from epiphallar caecum; lack of penial flagellum; vagina longer than free oviduct; presence of large atrial appendage with extrudable conical-cylindrical stimulator. The closest match appears to be with Ratnadvipia Godwin-Austen, 1899, a Sri Lankan endemic genus placed in Ariophantidae by SCHILEYKO (2002b: 1287), with a second species described by RAHEEM & NAGGS (2006). This would be a zoogeographically improbable ally, especially since both Ratnadvipia species have a much larger shell (B >18 mm), the bursa copulatrix lacking a separate duct, a much longer sarcobelum [atrial

39

appendage] and apparently different penis structure, so they seem more likely to show convergent affinities rather than be closely related to Principicochlea. Only the type species Principicochlea tenuitesta from Príncipe is referred to this genus, since Vitrina dumeticola Dohrn is clearly different, as discussed above. The type material and only specimens known of P. tenuitesta were collected from leaves in foliage above the ground in low mossy forest at 860 m alt. high on Pico Príncipe, which at 947 m forms the highest point of the island. Further research is needed to establish the extent of its range on the mountains of the island. Attempts to refind Vitrina dumeticola in the lowland Azeitona of north Príncipe are also needed in order to collect anatomical specimens to establish its taxonomic position.

Principitrochoidea gen. nov. D. Holyoak & G. Holyoak

Type species: Helix Folini Morelet, 1848 (MORELET, 1848: 352).

Etymology: Principi- is derived from the name of the island of Príncipe; Brown, 1827 is the name of a genus of land snails (Geomitridae) that are similar in having the shell shaped like a spinning top toy. Like Trochoidea, the new generic name is a feminine noun.

The new genus contains three species, P. aglypta, P. folini and P. convexa sp. nov. All three occur on the island of Príncipe, reports from elsewhere being only of P. folini from Angola, Gabon and Cameroon (GERMAIN, 1912b: 353-354) and Bioko (GASCOIGNE, 1994a: 2; WRONSKI ET AL., 2014: 165, table 1). Separation of Principitrochoidea from other genera of Helicarionoidea is based mainly on characters of the distal genital anatomy, which has not been described previously. The literature is also unclear regarding the shell characters useful for identifying the species. We therefore give separate accounts of the bodies and genital anatomy of each species here, ending with a key for species identification based mainly on shells.

Principitrochoidea convexa sp. nov. G. Holyoak, D. Holyoak & F. Sinclair (Figs 3C, 9E- G)

Type material: Holotype, NHMUK reg.no., shell (H 6.83, B 9.51 mm) + bod, collected 3 Dec. 2018, DTH, GAH & FS, at site 18-11. Type locality: Príncipe Island, just S. of Santo António on road to Bela Vista, N1.6307º E7.4179º, ca 24 m alt., edges of secondary forest, trackside banks & cultivation. Paratypes: All from Príncipe Island: from type locality (site 18-11) 3 sh; 4 Dec. 2018, path to Santa Joaquina, N1.6157° E7.3976°, 216 m alt., tall native forest with understorey of saplings, found on leaf-litter at base of steep flushed bank, GAH, DTH & FS 18-12, 12 sh (including 3 imm.) + 4 spm; 4 Dec. 2018, just S. of Santo António on road to Bela Vista, N1.6300º E7.4177º , ca 24 m alt., banana cultivation, edges of secondary forest by track, GAH 18-14, 9 sh (including 2 old & 1 imm.), 1 spm; 5 Dec. 2018, 1.5 km SW of Terreiro Velho, close to Oquê Pipi, N1.59965° E7.41082°, 308 m alt., tall native forest, GAH 18-16A, 6 sh (including 2 imm.) 1sh + bod; 3 May 2019, Floresta de Ponta do Sol, N1.65121º E7.39728º, 198 m alt., tall secondary forest, leg. A. Pedronho & R. Fernandes 2019-05-03-1, 1 sh; 15 May 2019, Morro de Leste trail 2, N1.59631º E7.41166º, 375 m alt., tall disturbed native forest, leg. A. Pedronho, A. Andrade & R. Fernandes 2019-05-15-2, 3 sh. Other material examined: From Príncipe Island, 4 Dec. 2018, path to Santa Joaquina, N1.6131º E7.3968º, 240 m alt., tall native forest with understorey of saplings (sampled by sieving), GAH, DTH & FS 18-13B, 3 sh (imm.).

40

Etymology: The epiphet convexa is based on convexus (Latin, meaning convex or curved outwards), referring to the shell being more convex beneath than in the two congeneric species.

Description: Based on holotype and paratypes (with measurements only from mature shells), H 6.63-7.44 mm, B 9.11-9.92 mm, H/B 0.72-0.75; AH 4.49-5.09 mm, AB 4.76-5.03 mm, AH/AB 0.94-1.03; UB 0-0.34 mm (UB/B 0-3.5%); whorls 5.4-6.2. Shell dextral, above slightly depressed conical with pointed spire, below convex. Protoconch small; whorls increasing gradually at first, more rapidly in penultimate-whorl and body-whorl; whorls slightly rounded above; suture of moderate depth, appearing as narrow band or double line. Body-whorl slightly convex above periphery, deeply convex below it, with prominent peripheral keel above middle of whorl height; top edge of keel bordered with narrow concavity or flat area, resulting in low peripheral cord around outermost edge. Aperture almost round except where intersected by penultimate-whorl and at just above mid-height of palatal margin where keel forms angle of ca 120º with small internal sinus. Peristome thin, mainly plane, but end of columellar margin widely reflected at umbilicus. Umbilicus narrow and deep, most often fully overlapped by reflected peristome in mature shells, although usually visible in oblique view. Protoconch smooth or sometimes with faint spiral microsculpture. Teleoconch often with low, rather widely and evenly spaced, radial- tangential ribs which are almost straight; rib-crests in some shells with small curved hairs up to 0.3 mm high; in other populations ribs weak or almost lacking. Shell surface between ribs with minutely decussate sculpture just visible at ×56 using strong oblique illumination. Underside of shell smooth except for very low irregular radial riblets. Shell thin, rather fragile, translucent; colour light brown to brown; waxy lustre above, glossy below. Body elongate with long tail; caudal-horn long (2.5 mm), tapering apically to slender point that extends backwards beyond end of body; end of body with deep caudal fossa with large diamond-shaped external opening. Mantle-collar a broad flap extending around most of periphery of body, widest anteriorly. At front edge its lower part is thicker and muscular, upper part thinner, the latter ending at each side in mantle-laps directed backwards. The right mantle-lap starts in a more posterior position than the left mantle-lap, but they are variable between individual snails in length (2-3 mm) and either right-lap or left-lap may be widest. Exterior of body mainly whitish, but pale brown on dorsum of front end and caudal-horn light grey. Mantle-laps whitish; mantle inside shell very pale brown with scatter of large white spots on right-hand half; body inside middle of spire blackish-brown; digestive-gland in upper spire light brown. Genital pore located just below mid-height towards front of right flank of body, slightly in front of base of upper tentacle; its external lips slightly swollen around the concealed pore. Genital atrium a short tube, dividing proximally on one side into distal penis, on opposite side into distal end of free oviduct, with exit from large duct of bursa copulatrix in between these organs. Penis large, its distal portion tubular, widening at distal end of longer proximal portion, which then tapers proximally to long slender terminal caecum that ends in slender retractor muscle attaching to proximal end of spermoviduct nearby. Epiphallus and flagellum lacking. Longitudinal section of penis (Fig. 9G) reveals distal portion with single muscular wall, longer proximal portion with double muscular wall (inner wall joined to outer wall by many strands of connective tissue); interior surface of (inner) wall of most of penis with longitudinal muscular pilaster, largest in distal portion where semi-circular in cross-section, becoming lower and narrower proximally as it extends along most of length of lumen of proximal penis. Vas deferens starts near distal end of spermoviduct, passing distally alongside free oviduct to distal end of duct of bursa copulatrix and near distal end of distal penis, returning proximally alongside penis and entering tip of penis at base of its terminal

41

caecum. Vas deferens a ± narrow tube throughout almost all of its length, but at level of distal penis a prominent bulge somewhat longer than wide begins and ends rather abruptly, its thickness about double that on each side. The bulge present in both specimens studied; the most mature specimen also had vas deferens thickened and convoluted alongside proximal penis, with conspicuous white contents. Vagina lacking. Free oviduct a rather thin tube only about half length of penis, its distal section slightly wider with a thin translucent external wall, its proximal section thinner but with a thicker wall. Duct of bursa copulatrix wide and thick-walled, tapering proximally, nearly as long as penis; reservoir of bursa copulatrix wider than adjacent part of duct, in one specimen joined apically to proximal end of prostatic part of spermoviduct by thin strands of tissue; in other specimen not obviously joined and reservoir swollen (with remains of spermatophore?). Spermoviduct rather short (wide part only two-thirds length of penis, but with shorter and more slender proximal section); albumen-gland short, arcuate; common hermaphrodite duct a narrow tube (broken-off proximally). Right ommatophore retractor passes between distal penis and duct of bursa copulatrix. Comments: : The terminology of organs of the penis complex in this genus is problematical, partly because their functions are uncertain. The proximal continuation of the penis is described as a "terminal caecum" here because it appears to arise directly from the penis and a small muscle attached to its most proximal part is likely to function as a penial retractor. The alternative of terming it an epiphallus is avoided because the vas deferens inserts at or just beyond its distal end. Nevertheless, it is possible that the tail of the spermatophore forms within this "terminal caecum" and if so the term "flagellum" might be more appropriate, although a retractor muscle arising apically would not be expected when that term is used. P. convexa shares the distinctive characters of P. folini which support separation of Principitrochoidea from Trochonanina. However, the two species differ considerably in the internal structure of the penis, with a distal verge and small lateral caecum present in P. folini, whereas P. convexa sp. nov. lacks both but has a muscular longitudinal pilaster on the inner wall of the penis lumen. Our records of P. convexa were from a wide variety of habitats at 24-375 m alt., including gardens with banana plants and bushes, secondary forest and its edges and (repeatedly) in tall native forest. At least some of the living individuals were collected from foliage above the ground. At site 18-12 it was living in forest with both P. aglypta and P. folini. The Key below differentiates P. convexa from congeners. Since this species was not reported by collectors active on Príncipe during the nineteenth century, it is unclear whether it was overlooked by them or it has arrived on the island since then as an introduction. Its occurrence at several sites in native forest far from human habitation may imply that it was previously overlooked, or if it is not native, that it has spread widely.

Principitrochoidea folini (Morelet, 1848) (Fig. 9H, I)

Helix Folini Morelet, 1848, Rev. Zool., 11, p. 352; TL insulam Principis sinus Guineensis. Syntypes were formerly in NHMUK, but they were not re-located by BREURE ET AL. (2018: 287). Trochonanina percarinata Martens, 1876, Monatsb. Königl. Akad. Wiss. Berlin, p. 256, pl. 1, figs 16-18; TL Bonjongo [im Camerun-Gebirge]. Trochonanina (Trochozonites) Folini Morelet: GERMAIN (1912b: 353-354), giving records from Príncipe from Roca Infante D. Henrique, 100-300 m alt.; Bahia do Oeste, 100-200 m

42

alt.; in footnote 2, he treated taxon described by MARTENS (1876) as Trochonanina (Trochozonites) Folini var. percarinata Martens. Trochonanina aglypta (Dohrn, 1866) sensu COSSIGNANI (2014: 194), non Morelet, 1848 (error of identification; the locality "São Tomé" given by Cossignani is presumably also erroneous). Trochonanina folini (Morelet, 1848): BREURE ET AL. (2018: 287, fig. 417).

Material examined: All from Príncipe Island, 3 Dec. 2018, just S. of Santo António on road to Bela Vista, N1.6307º E7.4179º, ca 24 m alt., edges of secondary forest, trackside banks, cultivation, GAH, DTH & FS 18-11, 1 sh + bod; 4 Dec. 2018, path to Santa Joaquina, N1.6157° E7.3976°, 216 m alt., tall native forest with understorey of saplings, found on leaf- litter at base of steep flushed bank, GAH, DTH & FS 18-12, 6 sh (including 2 old & 3 imm.); 4 Dec. 2018, just S. of Santo António on road to Bela Vista, N1.6300º E7.4177º, ca 24 m alt., banana cultivation, edges of secondary forest by track, ca 24 m alt., GAH 18-14, 9 sh (including 2 old & 1 imm.), 1sh + bod, 1 spm; 5 Dec. 2018, by road N. of Santo António, N1.64506º E7.41998º, 58 m alt., by plantation of bananas, etc., GAH 18-17, 2 sh (old).

Description: Compared to the somewhat larger P. convexa sp. nov. described in more detail above, the caudal-horn of P. folini is longer (3.5 mm). The arrangement of its mantle- collar and laps is similar, although the mantle-collar is less wide and the mantle-laps are shorter (right-lap 1.7 mm, left-lap 1.3 mm). Its body coloration is darker, with exterior of body light grey, shading to pale grey on foot fringe and dorsum of rear part of body beneath shell; tail horn is grey. Mantle-collar grey; mantle surface inside shell (sample 18-14) pale grey-brown with only small and faint dark markings or (sample 18-11) grey with minute blackish speckling. Genital anatomy (Fig. 9H, I) has similar general structure to that of P. convexa described in more detail above. Its penial terminal caecum is even longer in proportion to penis length, when in situ resting against spermoviduct; proximal tip of terminal caecum missing in one specimen and likely to be incomplete in other, so no muscular attachment seen. Terminal caecum thinner-walled than penis. Penis large with narrower distal portion forming only one-fifth of total penis length; proximal portion mainly cylindrical, but tapering proximally into terminal caecum. Longitudinal section of penis studied in one specimen (Fig. 9I) reveals thin muscular outer sheath along full length of proximal penis, continuing as thinner membranous sheath over distal penis. Inside sheath, central axis formed by thick muscular-walled tube running almost full length of penis, with small rounded lateral caecum directed backwards just distal of the middle. Middle part of the tube free of penis sheath, but proximal one-third of it and the distal end firmly attached to inside of sheath by connective tissues; tube widening at its distal extremity, apparently forming a verge. Thus penis probably partly eversible with verge forming intromittent organ. Vas deferens stout and convoluted in section adjacent to proximal penis; with white "globular process" on section near distal penis. Bursa copulatrix larger, ovate, when in situ appressed to prostatic side of spermoviduct. Comments: Principitrochoidea folini shows several clear differences in its distal genital anatomy from Trochonanina mozambicensis (L. Pfeiffer, 1855), the type-species of Trochonanina, described and figured by SCHILEYKO (2002b: 1278-1279, fig. 1682), which have led us to name a new genus. Thus, T. mozambicensis has a moderately long vagina, a distinct epiphallus, a short free flagellum and a large penial gland, all lacking in P. folini. Nevertheless, they share the peculiar swelling in the middle part of the vas deferens (termed "globular process on middle portion" by Schileyko), lacking so far as known in other

43

helicarionoids. Along with the shared trochoid shell shape, this character is likely to suggest affinity of these genera. P. folini was recorded by us from 24-216 m alt., mainly in disturbed habitats and cultivated areas, but also in numbers at a site in tall native forest with understorey of saplings, away from human habitation. At the latter locality, it was living near both P. aglypta and P. convexa. Since the species is well known in neighbouring countries of the African mainland as well as on Bioko, it might have originally been introduced to Príncipe, despite its original discovery on the island being in 1848 or earlier.

Principitrochoidea aglypta (Dohrn, 1866)

Nanina aglypta Dohrn, 1866, Malak. Blätt., 13, pp. 119-120, pl. 5, figs 5-7; TL Ilha do Principe in "silvis montosis ad Lappa d'Oeste". Treated as Trochonanina aglypta Dohrn by Germain (1912b: 340) and Trochonanina (Trochozonites) aglypta Dohrn by Germain (1916: 162).

Material examined: All from Príncipe Island, 4 Dec. 2018, path to Santa Joaquina, N1.6157° E7.3976°, 216 m alt., tall native forest with understorey of saplings, found on leaf- litter at base of steep flushed bank, GAH, DTH & FS 18-12, 3 sh + 1 bod; 15 Apr. 2019, Pico Príncipe campsite, N1.59665º E7.37301º, 265 m alt., tall native forest, leg. Y. dos Santos 2019-04-19-2, 2 spm; 3 May 2019, Floresta de Ponta do Sol, N1.65121º E7.39728º, 198 m alt., tall secondary forest, leg. A. Pedronho & R. Fernandes 2019-05-03-2, 1 sh; 15 May 2019, Morro de Leste trail 2, N1.59967º E7.41085º, 344 m alt., tall disturbed native forest, leg. A. Pedronho, A. Andrade & R. Fernandes 2019-05-15-3, 1 spm. Description: Largest specimen drowned and preserved in alcohol has partly hunched body 9.5 mm long; another smaller but less hunched specimen preserved in the same way has body 12.5 mm long. Caudal-horn is proportionately very long (3.3 mm on largest body, 4.2 on smaller body), conical basally with filiform tip 1.4 mm long, erect, arched forwards, but it would project considerably beyond tail tip if lowered. Beneath the caudal-horn the caudal fossa is evident as a round hole on the end of the tail. Mantle-laps proportionately long, very elongate triangular tapering to sharp point, in largest specimen right-lap 2.4 mm, left-lap 3.5 mm with separate 2.0 mm process conjoined to outer edge. External parts of body mainly light grey, shading to pale grey on foot-fringe; caudal horn grey; dorsum whitish beneath shell on base of tail and on sole of foot; dorsum of front of body blackish, with blackish ommatophore retractor muscles showing through translucent skin. Mantle-collar and flaps light grey with whitish middle band. Mantle surface inside shell pale but with blackish closely spaced minute spiral lines giving greyish effect overall. External body-wall of whorls inside spire with blackish spiral lines. An apparently immature specimen has genital pore on right right-hand side of body, one- third up from base of foot below base of upper tentacle. Genital anatomy studied in single specimen which nearly mature but with damage around head and genital atrium. Penis large and long, with structure ± as in P. folini; membranous sheath lost from distal end where verge with slit-like end pore remains intact; terminal caecum nearly as long as penis, curved, cylindrical, thin, when in situ attached to prostatic side of spermoviduct, with fine retractor muscle arising at its proximal tip and passing distally to join underside of proximal part of spermoviduct. Vas deferens passes distally as thin tube alongside free oviduct, with white "globular process" level with distal penis, becoming thick and convoluted where passing alongside inner edge of penis, inserting at tip of penis. Vagina nil or almost nil, with free oviduct arising as thin tube alongside thicker tube of bursa copulatrix duct from (remains of) genital atrium beside distal end of penis. Bursa copulatrix large and ovoid with thin wall.

44

Comments: DOHRN (1866b: 119-120, pl. 5, figs 5-7) described this species on the basis of a single shell containing a dried body found by a local helper. There have been no subsequent reports of it being collected. Our discovery of a third species of the genus on Príncipe necessitates critical reassessment of the identity of Dohrn's material. This is made more difficult because his collection was formerly housed in the Museum at Stettin (now Szczecin, Poland) but totally destroyed in the 1939-45 war (DANCE, 1986: 210). Fortunately, Dohrn made it clear that his shell was as high as wide with aperture wider than high ("Diam. maj. 10, min. 9, alt. 10, apert. lat. 5, alt. 4 mill.") and his figures confirm these proportions. He also described the sculpture as "radiato-striatulus" and followed the account of N. aglypta with comments on newly collected Nanina Folini, pointing out that the latter had the suture "vielmehr canalförmig" with a weak spiral rib beneath, and thus making it clear that the two species were different. The dissection of our single damaged subadult specimen of P. aglypta shows enough of its genital anatomy closely resembles that of P. folini to confirm the place of this species in genus Principitrochoidea. Furthermore, it resembles P. folini rather than P. convexa sp. nov. in having a well-developed distal verge in the penis. Unlike the other two species we place in this genus, P. aglypta is known only from forest habitats, at 198-344 m alt. All of our four records were from tall forest, three of them noted as native forest (disturbed forest at one locality) the fourth as secondary forest. At site 18-12 it occurred with both P. folini and P. convexa sp. nov. The only previous record was of the type specimen, recorded as from "silvis montosis" (mountain forest).

Key to species of Principitrochoidea based mainly on shell characters

1. Upper edge of body-whorl just below suture with concave channel bordered below by spiral rib; periostracum of body-whorl with coarse sculpture of radial ribs bearing stiff erect hairs, the longest of them >0.5 mm long P. folini

-. Upper edge of body-whorl just below suture often with narrow flatter strip but lacking concave channel bordered by spiral rib; periostracum of body whorl smooth or with sculpture of radial ribs; if erect hairs present these <0.5 mm long 2.

2. Shell wider than high (H/B 0.7-0.8); base of shell usually more convex (so keel in apertural view normally on upper half of peristome); body-whorl smooth above or with ± strong radial ribs; penis lacking distal verge (Fig. 9G) P. convexa sp. nov.

-. Shell about as wide as high (H/B 0.8-1.1); base of shell less convex (so keel in apertural view usually around middle of peristome or on its lower half); body-whorl with sculpture of fine radial riblets; penis with distal verge P. aglypta

Thomithapsia gen. nov. D. Holyoak & G. Holyoak

Type species: Thomithapsia bomsucessica sp. nov. Etymology: Based on Thomi- from an old name (île de Saint Thomas) for São Tomé island and snail genus Thapsia Albers, 1860 of Helicarionidae (see comments above under Apothapsia regarding apparent Greek origin of Thapsia). Like Thapsia, it is a feminine noun.

Description: the new genus is based only on the following species:

Thomithapsia bomsucessica sp. nov. G. Holyoak & D. Holyoak (Fig. 3D, 8G-J)

45

Type material: Holotype NHMUK reg.no., shell (H 6.97, B 12.33 mm) & bod, collected 4 Dec. 2013, by GAH & DTH at site ST4. Type locality: São Tomé Island, path to Lagoa Amélia from Bom Sucesso Botanical Garden, N0.2786° E6.6000°, ca 1300 m alt., montane forest on slope with understorey of saplings. Paratypes: All from São Tomé Island: 1 sh (dead when found) from type-locality; 3 Dec. 2013, on path towards Macambrará, near Rio Manuel Jorge, N0.2847° E6.6097°, ca 1149 m alt., trackside, cultivation edges, scrub & small forest patches, GAH & DTH ST3, 1 sh; 24 Nov. 2018, Morro Claudina (near Bom Sucesso), N0.2912° E6.6057°, ca 1289 m alt., montane rainforest on ridges & slopes, GAH, DTH, et al., 18-2, 3 spm; 24 Nov. 2018, near Quinta da Graça, N0.29928° E6.64174°, 678 m alt., scrub, gardens, roadside banks, with few trees, GAH & DTH 18-3, 1 sh; 27 Nov. 2018, Morro Claudina (near Bom Sucesso), N0.2918° E6.6059°, 1254-1292 m alt., montane rainforest, GAH & DTH 18-6, 1 sh & bod; 9 Dec. 2018, path to Lagoa Amélia from above Bom Sucesso Botanical Garden, N0.2823° E6.5964°, 1290-1415 m alt, montane forest, GAH & DTH 18-21, 2 sh & 2 spm. Etymology: The epiphet bomsucessica is an adjective based on the name of Bom Sucesso (in distrito de Mé-Zóchi), on São Tomé island, the location of the famous Botanical Garden and close to the collection localities of most specimens of the new species.

Description: Based on holotype and paratypes (with measurements only from mature shells), H 6.97-8.31 mm, B 12.33-13.33 mm, H/B 0.57-0.64; AH 5.03-5.90 mm, AB 5.90- 6.70 mm, AH/AB 0.85-0.91; UB 0.27-0.47 mm (UB/B 2.2-3.5%); whorls 5.2-5.7. Shell dextral, depressed-conical with rounded spire and flatter base. Protoconch small; whorls increasing gradually, with body-whorl expanding more rapidly to rather wide aperture; suture moderately deep, appearing as narrow band or double line. Body-whorl rounded at periphery, with slight hint of weak keel on upper edge of periphery that is much more obvious in immature shells. Aperture broadly oval except where interrupted by penultimate whorl. Peristome thin and plane, narrowly reflected close to adjacent half of umbilicus; lower palatal margin with convex edge in ventral view. Umbilicus funnel-shaped above the narrow interior (where UB measured), deep, exposing spire internally, almost symmetrical, slightly overlapped by reflected peristome. Shell sculpture weak throughout; protoconch mainly smooth, with weak spiral microsculpture in places; upperside and underside of teleoconch with low irregular radial riblets and closely-spaced fine spiral grooves. Shell thin, rather fragile, slightly translucent. Protoconch and inside of umbilicus whitish, otherwise shell colour a rather bright light brown; periostracum glossy on upper whorls, with less gloss or waxy lustre on body-whorl. Body relatively long and slender; long tail ending dorsally in caudal horn (moderately tall, nearly erect, 2.0 mm high at front edge), with large caudal fossa in rear end beneath. Mantle-collar wide, with two small rounded lobes projecting forwards over fore-part of body (L1) and large projecting shell-laps on both sides of body (Fig. 8H). The left shell-lap (L3) ca 4 mm wide, with long loose flap having pointed tip directed posteriorly; the right shell-lap (L4) ca 5 mm wide, with long loose flap directed posteriorly having pointed tip and another round-tipped free flap extending outwards from anterior edge. The shell-laps are withdrawn inside shell when snails are approached closely or touched. Bodies of darker specimens preserved in ims have exposed parts of body pale grey above, whitish on foot-fringe and sole of foot, mantle-laps pale to light grey dorsally, dorsal surface of mantle inside shell dark grey to nearly blackish with pale lower fringe. Bodies of lighter specimens are mainly whitish externally, shading to very pale brown high on flanks and on mantle-laps, to light brown on dorsum of fore-part and front of face; the ommatophore retractor muscles showing as dark grey through translucent skin. These have

46

mantle inside shell dorsally blackish-grey, its left-hand edge whitish; digestive gland inside spire dark brown. Genital anatomy studied in two mature individuals (sites ST4, ST18-6). Genital pore on right flank of forepart of body below base of right ommatophore at about one-third height, with surrounding tubercles whitish. Genital atrium a cylinder of medium length, its proximal end dividing into distal end of penis and distal end of vagina, with duct of bursa copulatrix inserting between these. A short but definite muscle also attached to exterior of wall of proximal end of atrium adjacent to base of bursa duct or base of distal end of penis. Penis moderately large and bulky, with shorter and narrower distal portion (one-third of penis length) extending proximally into longer and wider proximal portion (two-thirds of penis length), giving clavate outline. Short thinner penial caecum on proximal extremity of penis with penial retractor muscle arising from its end as long slender muscle that passes proximally through body cavity alongside other genitalia. Longitudinal section of penis (Fig. 8J) reveals rather thin muscular outer wall, the interior of which has two longitudinal muscular pilasters attached to it throughout most of length of proximal penis; another small rounded pilaster lies close to junction of proximal with distal penis; inner wall of distal penis has a fourth pilaster forming longitudinal ridge, larger than several other lamellae running parallel to it. Thus there is no distinct epiphallus, no penial flagellum and no verge inside penis. Vas deferens a slender tube throughout its length, begins near distal end of spermoviduct, passes distally alongside free oviduct into angle near bursa copulatrix duct and distal part of penis, before returning proximally alongside penis and entering wall of proximal penis close to its end. Vagina absent. Free oviduct slightly longer than penis but more slender, divided into thicker-walled distal part and thin-walled (translucent) proximal part, with constriction in between. Duct of bursa copulatrix short and narrow or somewhat wider, a tube, passing into small ovoid or long-ovoid bursa reservoir which is of approximately the same length. Spermoviduct as long as penis or somewhat longer; albumen gland not studied. Right ommatophore retractor muscle passes proximally between base of distal part of penis and base of duct of bursa copulatrix. Comments: The genus Thomithapsia recalls Vanmolia sjoestedti (d'Ailly, 1896) from Cameroon in having a shell with raised spire, tiny umbilicus and spiral periostracal sculpture combined with distal genitalia lacking a vagina and possessing a large clavate penis (DE WINTER (2008: 468-475). However, Thomithapsia differs: (a) in lacking a short penial epiphallus (its vas deferens enters the proximal penis wall directly), (b) in lacking a verge inside the proximal penis, and (c) having a smaller duct of the bursa copulatrix lacking the peculiar thickened walls of its counterpart in Vanmolia. Since the combination of its shell and anatomical characters appear to be unique (cf. SCHILEYKO, 2002b), we treat Thomithapsia as an independent monotypic genus, which is so far as known an endemic on São Tomé. T. bomsucessica is distinct conchologically from the other species of Helicarionoidea known from São Tomé in having adults that show a combination of a thin rather glossy shell, light brown rather than pale brown shell coloration, more than 5 shell whorls, and absence of an obvious peripheral keel (although immatures have keeled shells). It has been found in rather small numbers at sites from ca 678 m up to at least 1300 m alt., inside montane forest and in disturbed places at lower altitudes without woodland cover. The living specimens were obtained from beneath fallen wood and amongst leaf-litter and other debris on the ground. Unlike the much more common Apothapsia thomensis, none were found on vegetation above ground.

Thomitrochoidea gen. nov. D. Holyoak & G. Holyoak

47

Type species: Thomitrochoidea trindadensis sp. nov. D. Holyoak & G. Holyoak, see below. Etymology: Thomi- is derived from an old name (île de Saint Thomas) for the island of São Tomé; Trochoidea Brown, 1827 is the name of a genus of land snails (Geomitridae) that are similar in having a shell shaped like a spinning top toy. Like Trochoidea, the new generic name is a feminine noun.

Description: the new genus is based only on the following species:

Thomitrochoidea trindadensis sp. nov. D. Holyoak & G. Holyoak (Figs 3E, 9J, K)

Type material: Holotype NHMUK reg.no., shell (H 8.6, B 14.4 mm) + bod, collected 2 Dec. 2018, by GAH & DTH at site 18-10. Type locality: São Tomé Island, just S. of Milagrosa, on the road between Trindade and Bombaim, N0.27667° E6.65986°, 398 m alt., tall secondary forest on slopes, track edges with flushed rocky cuttings & abandoned cocoa plantations. Paratypes: All from São Tomé Island: 2 Dec. 2018, from type-locality at site 18-10, 2 spm, 42 sh found dead; 1 Dec. 2013, by EN1 well S. of Santa Catarina & inland of NW. coast, N0.2405º E6.4722º, ca 68 m alt., roadside banks, etc., in tall open secondary-forest on slope, GAH ST1, 1 sh + bod; 2 Dec. 2013, S. of EN2 and NW. of Ribeira Peixe, N0.9314° E6.5889°, ca 126 m alt., edge of tall secondary forest on hill above young oil palm plantation, GAH ST2, 1 sh + bod; 6 Dec. 2013, S. of EN2 and ca 0.4 km W. of Vila Aida, N0.0897° E6.5903°, ca 181 m alt., tall forest on ridge & slopes with understorey of saplings, GAH ST7, 2 sh (imm); 13 Dec. 2013, by EN2 just NE of Monte Mário "roça", N0.0606° E6.5561°, ca 236 m alt., forest on slopes by road with understorey of saplings & ferns beneath, GAH & DTH ST13, 3 sh + 2 bod; 11 Dec. 2018, by S. side of EN2 between the “roças” Fraternidade and Soledade (ca 1.6 km due S. of São João dos Angolares, centre), N0.1262° 6.6371°, 22 m alt., old cocoa plantations & bananas, under tall trees & palms, near stream (Água São Pedro), GAH & DTH 18-26, 1 sh + bod. Other material examined: 11 Dec. 2013, by EN2 ca 2 km S. of Monte Mário "praia", N0.0694° E6.5522, ca 101 m alt., roadside bank at foot of slope with secondary forest, GAH & DTH ST11, 1 sh (old); 12 Dec. 2013, by EN1 between Lagoa Azul [Morro Carregado is the name on the military maps] and Praia Mutamba, N0.4006° E6.6075°, ca 5 m alt., dead/drifted shells from soil surface & debris by small dry stream-course above shore in lower edge of valley with cover of dry woodland, GAH & DTH ST12, 2 sh; 1 June 2019, Cão Pequeno, Caué, N0.0996º E6.5306º, 200 m, native forest, RFL, 1 sh (adult, broken). Etymology: The epiphet trindadensis is adjectival, based on the name of the town of Trindade (Portuguese for the Trinity), which is near to the type-locality and the administrative centre of Mé-Zóchi District. Synonym: Trochonanina moreleti Germain, 1921 [sic] sensu COSSIGNANI (2014: 193), photographs of two shells from São Tomé, a 13 mm, b 14 mm, non T. moreleti Germain, 1915.

Description: Shell dextral, depressed-conical to very depressed conical with apex to spire rounded and underside moderately convex. When mature H 8.6-9.6 mm, B 14.4-16.4 mm, with 4.3-4.9 whorls. Protoconch small, whorls increasing gradually and regularly in width, except that body-whorl expands disproportionately towards aperture. Whorl-profile only slightly convex above, with shallow suture forming a narrow band or double line. All whorls keeled at periphery and body-whorl sharply angled with blunt-edged keel in middle of periphery; a slight concavity above the keel and an almost plane area just below it, but outer edge of keel not ending in a distinct cord. Aperture elongate-oval except where widely

48

interrupted by penultimate whorl, and externally with sharp angle of ca 70º inside the peripheral keel; peristome thin, plane except for reflection of columellar margin around the adjacent half of the umbilicus, the reflection extending slightly onto palatal margin of aperture. Umbilicus very narrow (ca 1 mm), deep, exposing spire internally, partly overlapped by peristome edge. Protoconch nearly smooth, teleoconch with weak sculpture of radial-oblique lines, low irregular riblets, or both; faint irregular spiral microsculpture just visible at high magnification (× 42) on parts of some shells, apparently absent on others. Shell rather thin, translucent when immature, less so when adult. Protoconch whitish to pale brown; teleoconch variable in colour between different localities: all shells in large sample from TL dull cream with single (mainly narrow, broader on few sh) spiral band of red-brown just above keel; second sample with single shell (ST1) was similar but without the dark band; four other small samples were of much darker shells (ST2, ST7, ST13, 18-26) with early whorls and first part of penultimate whorl dull cream, remainder including all of body-whorl chestnut or dull chestnut with irregular radial-oblique lines (between riblets) of cream, and cream outer edge to peripheral keel; underside of shell bright chestnut without markings, except white interior of umbilicus. Fresh shells are glossy beneath but they lack gloss above, except on and near protoconch. Body elongate with relatively long tail ending in short caudal horn that forms rounded tubercle on dorsal surface of tip, beneath which caudal fossa has large opening (triangular, with point on lower edge). Mantle-collar forms large flap continuous all around front and sides of body, except where deeply incised by slit at front of right-hand edge. In pale-bodied specimens from type locality (TL), sole of foot, foot-fringe and most of exposed part of body whitish, at front pale brown dorsally and on upper flanks, the ommatophore retractor muscles showing as dull dark grey-brown areas through the translucent skin. Exposed mantle-collar whitish; mantle inside body-whorl very pale brown with scattered small grey-brown spots and short longitudinal streaks, with a continuous dark line along the columellar edge. In darker-bodied specimens (site 18-26), exposed foreparts of body pale grey dorsally, mantle- collar pale grey, mantle surface inside body-whorl whitish to pale grey with small spots and irregular small longitudinal streaks of dull grey concentrated along columellar edge. Genitalia studied in single specimens from type-locality (Fig. 9J, K) and site ST2, another from site 18-26 had immature genitalia. External genital pore at mid-height on right- hand side of forepart of body, below back edge of base of right upper tentacle. Genital atrium a short cylinder, dividing proximally into distal penis on one side and distal end of free oviduct on opposite side, with bursa copulatrix inserting between them. Penis clavate proximally, with two distinct portions: the distal penis a short tube, the proximal penis nearly twice as wide and twice as long and sharply bent just above its junction with distal penis. Penial retractor muscle inserted on proximal end of outer wall of penis, extending proximally alongside spermoviduct to deep within body cavity, apparently ending amongst columellar musculature. Penis opened longitudinally in one mature specimen (Fig. 9K), revealing a moderately thick muscular outer wall throughout; almost the full length of the inner wall of proximal penis with two high curved raised longitudinal lamellae (pilasters), both extending into distal penis and becoming lower there before ending in its distal half; within proximal penis the crests of the lamellae extend across more than half of the width of the lumen. Penial caecum, epiphallus and flagellum all absent; no trace of external glandular tissue around penis. Vas deferens a thin tube beginning near distal end of spermoviduct, passing distally alongside free oviduct, then in turn passing close to bursa copulatrix and into angle near base of distal penis, returning proximally alongside penis, finally entering proximal wall of penis just distal to insertion of its retractor muscle. Vagina lacking; free oviduct a tube about as long as the penis. Bursa copulatrix narrowly ovoid, rather small (only about half length of penis and much narrower than proximal penis), narrowed towards insertion onto genital

49

atrium but lacking a distinct duct, the outer wall of the proximal end giving rise to a long slender thread of strong tissue that passes proximally to join onto surface of proximal half of prostatic part of spermoviduct. Spermoviduct large, curved in situ; albumen gland moderately long, approximately ellipsoidal; common hermaphrodite duct a thin convoluted tube arising near distal end of albumen gland. Right ommatophore retractor passing proximally between distal part of penis and bursa copulatrix. No anatomical differences were detected between single mature specimens from populations consisting entirely of pale shells with a brown band (TL) and a population with dark shell coloration (site 18-26). Comments: Thomitrochoidea is described as a new genus here because we cannot match its characters with those of any other genus of Helicarionoidea. Although its shell characters recall those of Principitrochoidea gen. nov. from Príncipe described and discussed above, they do not match it closely and their distal genitalia show important differences. The type species of Trochonanina Mousson, 1869, T. mozambicensis (L. Pfeiffer, 1855), shows some similarity in shell characters, but differs markedly in having a distinct vagina and penis with a caecum, epiphallus, flagellum and large penial gland (SCHILEYKO, 2002b: 1278-1279), all lacking in Thomitrochoidea. PILSBRY (1919: 245) had presented similar anatomical data to those for T. mozambicensis for two other species from the Congo, placing them in the genus Ledoulxia Bourguignat, 1885 which is a synonym of Trochonanina. Plicatonina Verdcourt, 1961 has a very short vagina, but its penial caecum, epiphallus and duct of the bursa copulatrix are all well-developed (VERDCOURT, 1961). Other African helicarionoid genera with trochoid shells (Trochozonites L. Pfeiffer, 1883; Sjostedtina Verdcourt, 1961; Carinazingis van Bruggen & de Winter, 1990; Tropidocochlion Verdcourt, 1998) likewise differ in having an epiphallus and flagellum, among various other distinctive characters (SCHILEYKO, 2002b). Based on recent records from nine localities on São Tomé at 22-398 m alt., T. trindadensis sp. nov. appears to be restricted to the lowlands at sites with woody vegetation, including secondary-forest and its edges and old cocoa plantations. Except at the type locality, only a few specimens were found at each site and the living ones were mainly picked off vegetation above the ground. The type-locality is in an old cocoa plantation with tall Artocarpus trees where we initially collected 42 mainly fresh shells found scattered on the ground. Sustained searching did not reveal any living on or near the ground, so we examined saplings by bending them over to check their upper leaves and branches and this technique yielded three live specimens from 2.3 m above ground which were resting together inside an empty, decaying cocoa Theobroma cacao "pod" (fruit). Hence, it seems that the species might previously have been overlooked on the island because of its arboreal habitat. Since it has moderately large shells that were perhaps unlikely to have been overlooked by the nineteenth-century collectors, there is also a strong possibility that the species is a recent introduction to São Tomé from elsewhere in tropical Africa, where there are numerous poorly known nominal species of Helicarionoidea with trochoid shells. Nevertheless, the wide scatter of our localities from the northern to the southern end of the island would imply it may not have arrived recently. Also, there has been one find of it in a remnant of native forest at low elevation (1 June 2019, Cão Pequeno, N0.0996º E6.5306º, in native forest at 200 m alt., a fresh broken shell, leg. RFL).

HELICIDAE Rafinesque, 1815 Subfamily Helicinae, Tribe Otalini Pfeffer, 1830 Cornu Born, 1778

Cornu aspersum (O.F. Müller, 1774)

50

Helix aspersa O.F. Müller, 1774, Hist. Verm. terr. fluv., p. 59; TL Italy. New to São Tomé: 3 Dec. 2013, ca 1.5 km E. of Botanic Gardens at Bom Sucesso, ca 1100 m alt., among grass & weeds by track, GAH & DTH [E. of ST3], 1 typical living individual ca two-thirds adult size recorded but not collected; 24 Nov. 2018, by Bom Sucesso Botanical Garden, N0.28829° E6.61193°, 1159 m alt., trackside with herbs near hedges, scrub, wall & cultivation, GAH & DTH et al. 18-1, 2 sh & 1 bod. The species is now widespread in S. and W. Europe, where it has been regarded as native. Nevertheless, MADEC, BELLIDO & GUILLER (2003: 225) and GUILLER & MADEC (2010) concluded from evidence of loss of genetic variation and lack of convincing fossil evidence from Europe that the european populations result from historical introduction from NW. Africa by man. Elsewhere, it occurs eastwards to Turkey and the Middle East and is established as an introduction in S. Africa (widespread) and locally also in N. America, Mexico, Haiti, Argentina, Chile, Australia, New Zealand, Macaronesia and on St. Helena.

DISCUSSION

Table 1 presents the first comprehensive checklists since the work of GERMAIN (1916) for the terrestrial Mollusca of both islands, comprising 86 species in total, with 59 known for São Tomé and 45 for Príncipe. However, the faunas of both São Tomé and Príncipe are certainly not fully known yet. Our sieved material from forests on Príncipe includes two tiny land-shell apices that definitely do not represent species known on the island, one probably of Cerastidae, the other possibly , but the material available is so poor that it would be unhelpful to future researchers to give names to such scraps. Among the species receiving new names here, five are as yet known only from the unique type specimens, two from montane forest on São Tomé and three from Príncipe. Furthermore, our existing knowledge for São Tomé relies on intensive study at very few sites which happen to combine species-rich habitats and relatively easy access, notably the Bom Sucesso to Lagoa Amélia path and neighbouring Morro Claudina forest. Wider searching on both islands will doubtless reveal more undescribed species. The recent declines of some large endemic snails, including Archachatina bicarinata (DALLIMER & MELO, 2010) and Columna columna (pers. obs.) emphasises that the native faunas of these islands are not safe and static, while deleterious habitat changes and more introductions of alien species are occurring. As elsewhere on oceanic islands (CHIBA & COWIE, 2016; YEUNG & HAYES, 2018), it is urgent to find out more before the opportunity to do so is lost forever as endemic species become extinct. More taxonomic knowledge is still needed to inform effective conservation.

ACKNOWLEDGEMENTS

We are grateful for energetic practical assistance from the Fundação Príncipe, especially colleagues of FS in the fieldwork team (Aramis Andrade, Ayres Pedronho, Davide Dias, Yodiney dos Santos), whose work was supported by a grant from the French Facility for Global Environment (FFEM). The fieldwork in São Tomé had the support of BirdLife International, through the ECOFAC6 and Rainforest Trust (4875) projects. RFL and MP were supported by the Portuguese Government through the "Fundação para a Ciência e a Tecnologia" (FCT/MCTES – SFRH/BPD/91494/2012 & PD/BD/140814/2018, respectively), which also funded this work within the framework of the project UID/BIA/00329/2019. MP also received support from Critical Ecosystem Partnership Fund (109607). Permits to collect and export specimens were provided by Eng. Arlindo de Carvalho, Director Geral of Ministério das Infraestructuras Recursos Naturais e Ambiente, Direcção Geral do Ambiente

51

of the Republic of São Tomé and Príncipe. Rui Mendes assisted in obtaining literature. Tom Blockeel kindly advised us on formation and orthography of the new scientific names.

BIBLIOGRAPHY

ADAM W. 1954. Études sur les mollusques de l'Afrique centrale et des regions voisines 1. et . Volume Jubilaire Victor van Straelen. Tome II. Institut Royal d'Histoire Naturelle de Belgique. Pp. 725-817. ADAM W. 1957. Études sur les mollusques de l'Afrique centrale et des régions voisines 1. Vertiginidae et Valloniidae (addendum). Bulletin de l'Institut royal des sciences naturelles de Belgique, Biologie, 33: 1-26. ADAMS C.B. 1845. Specierum novarum conchyliorum, in Jamaica repertorum, synopsis. Proceedings of the Boston Society of Natural History, 2: 1-17. ANCEY C.F. 1885. Nouvelles contributions malacologiques. Bulletins de la Société Malacologique de France, 2: 113-146. BENSON W.H. 1850. Characters of new species of Helix from India, Mauritius and the Cape of Good Hope; also of a new Mauritian Tornatellina, with remarks on the habits of a Cape Succinea. Annals and Magazine of natural History, (2) 6: 251-256. BEQUAERT J. & CLENCH W.J. 1936. Studies of African land and fresh-water mollusks VIII. New species of land operculates, with descriptions of a new genus and two new subgenera. Revue de Zoologie et de Botanique africaines, 29: 97-104. BOUCHET P., ROCROI J.-P., HAUSDORF B., KAIM A., KANO Y., NÜTZEL A., PARKHAEV P., SCHRÖDL M. & STRONG E.E. 2017. Revised classification, nomenclator and typification of gastropod and monoplacophoran families. , 61 (1-2): 1-526. https://doi.org/10.4002/040.061.0201 BRANDT J.F. VON 1835 in: Recueil des actes de la séance publique de l'Académie impériale des sciences de Saint-Pétersburg, 1835: 250. n.v. BREURE A.S.H., AUDIBERT C. & ABLETT J.D. 2018. Pierre Marie Arthur Morelet (1809- 1892) and his contributions to malacology. Nederlandse Malacologische Vereniging, Leiden. 544 pp. BRUGGEN A.C., VAN 1986. Further notes on Afrotropical land molluscs (Gastropoda Prosobrancia: Maizaniidae, Cyclophoridae). Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, (C) 85: 179-204. BRUGGEN A.C., VAN 1994. The genus Truncatellina Lowe, 1852 (Mollusca, Gastropoda : ) in Malawi, Zambia, Angola, and Fernando Poo. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, (C) 97: 1-25 (with Erratum, pp. 377-378). BRUGGEN A.C., VAN & VAN GOETHEM J.L. 2001. Prolegomena for a checklist of the terrestrial molluscs of the Upemba National Park, Katanga, D.R. Congo. Bulletin de l'Institut royal des sciences naturelles de Belgique, Biologie, 71: 151-168. BRUGGEN A.C., VAN & DE WINTER A.J. 1995. Notes on Micractaeon, a monotypic genus of African land snails (Gastropoda Pulmonata: Ferussaciidae?). Zoologische Mededelingen, Leiden, 69 (7): 79-92. CHIBA S. & COWIE R.H. 2016. Evolution and extinction of land snails on oceanic islands. Annual Review of Ecology, Evolution and Systematics, 47: 123-141. DOI: 10.1146/annurev-ecolsys-112414-054331 COSSIGNANI T. 2014. African landshells. Atlas. L'Informatore Piceno, Ancona, Italy. 207 pp. COWIE R.H. 1998. Catalog of nonmarine snails and slugs of the Samoan Islands. Bishop Museum Bulletin in Zoology, 3: i-viii, 1-122.

52

COWIE R.H., DILLON R.T., ROBINSON D.G. & SMITH J.W. 2009.Alien non-marine molluscs of priority quarantine importance in the United States: A preliminary risk assessment. American Malacological Bulletin, 27: 113-132. CROSSE H. 1868. Faunule malacologique terrestre de l'île San-Thome. Journal de Conchliologie, (3) 8: 125-135, pl. 6 figs 2, 5. CROSSE H. 1888a. Nouveau catalogue des mollusques terrestres de l'ile de San-Thomé. Journal de Conchliologie, (3) 28 (1): 13-30, pl. 1 fig. 2, 2a. CROSSE H. 1888b. Faune malacologique terrestre et fluviatile de l'île du Prince (Côte occidentale d'Afrique). Journal de Conchliologie, (3) 28: 296-305. CROWLEY T.E. & PAIN T. 1970.A monographic revision of the African land snails of the genus Limicolaria Schumacher (Mollusca – Achatinidae). Annales du Musée Royal de l'Afrique centrale, Tervuren [Sciences zoologiques], 177: 1-61, 6 pls. DALL W.H. 1905. Note on Trichodina Ancey. Nautilus, 18 (12): 143. DALLIMER M. & MELO M. 2010. Rapid decline of the endemic giant land snail Archachatina bicarinata on the island of Príncipe, Gulf of Guinea. Oryx, 44 (2): 213-218. DANCE S.P. 1986. A history of shell collecting. E.J. Brill – Dr. W Backhuys, Leiden. Pp. xv + 265, 32 pls. DOHRN H.1866a. Eine neue Nanina aus Westafrika. Malakozoologische Blätter, 13: 114-115. DOHRN H.1866b. Die Binnenconchylien von Ilha do Principe. Malakozoologische Blätter, 13: 116-136, pl. 5. EMBERTON K.C., SLAPCINSKY J., CAMPBELL C.A., RAKOTONDRAZAFY J.A., ANDRIAMIARISON T.N. & EMBERTON J.D. 2010. Terrestrial mollusks of Andriantantely Massif, Eastern Madagascar, with descriptions of 36 new species (Gastropoda: ; Pulmonata). Archiv für Molluskenkunde, 139 (1): 71-141. https://doi.org/10.1127/arch.moll/1869-0963/139/071-141 FERNANDES F. & ROLÁN E. 1993. Moluscos marinos de São Tomé y Principe: actualización bibliográfica y nuevas aportaciones. Iberus, 11(1): 31-47. FÉRUSSAC A.E.J.P.F. D'AUDEBARD DE 1821-1822. Tableaux systématiques des animaux mollusques classés en familles naturelles, dans lesquels on a établi la concordance de tous les systèmes; suivis d'un Prodrome général pour tous les mollusques ou fluviatiles, vivantes ou fossiles. Arthus Bertrand – Paris & G.B. Sowerby – London. Publication dates from KENNARD (1942): Livraison 9: 1-24 (Quarto edition) [Folio edition: 1-32] (6- IV-1821); livr. 10: 25-48 (Quarto) [Folio: 33-56] (26-V-1821); livr. 11: 49-72 (Quarto) [Folio: 57-76] (13-VII-1821); livr. 12: 73-88 (Quarto) [Folio: 77-92] (21-IX-1821); livr. 13: 89-110 (Quarto) [Folio: 93-114] (10-XI-1821); livr. 14: i-xxiv (Quarto) (16-II-1822); livr. 15: xxv-xlvii[i] (Quarto) (13-IV-1822); livr. 16: 1-27 (Quarto) (13-VII-1822). FORCART L. 1953. The Veronicellidae of Africa (Mollusca, Pulmonata). Annales du Musée Royal du Congo Belge, Tervuren (Sciences zoologiques), 23: 1-110, pls 1-5. FORCART L. 1954. Two species of Veronicellidae from the Kenya Colony and Afroveronicella nom. nov. for Hoffmannia Forcart. Proceedings of the Malacological Society of London, 31 (1): 20-21. GASCOIGNE A 1993. A bibliography of the fauna of the islands of São Tomé e Príncipe and the Island of Annobon (Gulf of Guinea). Arquipélago, Life and Marine Sciences, Angra do Heroísmo (Açores), 11A: 91-105. GASCOIGNE A 1994a. The dispersal of terrestrial gastropod species in the Gulf of Guinea. Journal of Conchology, 35 (1): 1-7. GASCOIGNE A 1994b. The biogeography of land snails in the islands of the Gulf of Guinea. Biodiversity and Conservation, 3: 794-807.

53

GASCOIGNE A 1996. Additions to a bibliography of the fauna of São Tomé e Príncipe and the Island of Annobon (Gulf of Guinea). Addendum. Arquipélago, Life and Marine Sciences, Angra do Heroísmo (Açores), 14A: 95-103. GERBER J. & CLARK S.A. 2015. First record of the predatory land snail Streptostele (Tomostele) musaecola (Pulmonata: Streptaxidae) in the continental United States. American Conchologist, 43 (4): 26-28. GERMAIN L. 1908. Mollusques terrestres recueillis par M. Ch. Gravier à l'île San Thomé (1906). Bulletin du Muséum national d'histore naturelle, Paris, 14: 55-62. GERMAIN L. 1912a. Descriptions de mollusques nouveaux de l'Ile du Prince (Golfe de Guinée) et de l'Afrique occidentale. Bulletin du Muséum national d'histore naturelle, Paris, 18 (5): 318-324. GERMAIN L. 1912b. Mollusques terrestres et fluviatiles recueillis par M. L. Fea pendant son voyage à la Guinée et à l'Ile du Prince. Annali del Museo Civico di Storia Naturale di Genova, (3) 5: 335-399, pl. 4. GERMAIN L. 1915. Contributions à la faune malacologique de l'Afrique équatoriale. XLI. Mollusques nouveaux des îles du Golfe de Guinée. Bulletin du Muséum national d'Histoire naturelle, Paris, 21 (7): 283-290. GERMAIN L. 1916. Etude sur les mollusques terrestres et fluviatiles recueillis par L. Fea pendant son voyage en Afrique occidentale et aux Iles du Golfe de Guinée. Annali del Museo Civico di Storia Naturale Giacomo Doria (ser. 3, vol. 7) 47: 150-336, pls. 6-11. GERMAIN L. 1934. Contributions à la faune malacologique de l'Afrique équatoriale. LXIX. Mollusques terrestres nouveaux des régions montagneuses de l'Afrique orientale (Mission de l'Omo, 1932-1933). Bulletin du Muséum national d'histoire naturelle, Paris, (2) 6: 377-382. GIRARD A.A. 1893a. Révision de la faune malacologique des Iles St. Thomé et du Prince. I.– Mollusques terrestres et fluviatiles. Jornal de Sciencias mathematicas, physicas e naturaes, Lisboa, (2) 3 (9): 28-42. GIRARD A.A. 1893b. Révision de la faune malacologique des Iles St. Thomé et du Prince. I.– Mollusques terrestres et fluviatiles (suite). Jornal de Sciencias mathematicas, physicas e naturaes, Lisboa, (2) 3 (10): 95-114, pl. 1. GIRARD A.A. 1895. Sur le "Thyrophorella thomensis", Greeff. Gastéropode terrestre muni d'un faux opercule à charnier. Jornal de Sciencias mathematicas, physicas e naturaes, Lisboa, (2) 4 (13): 28-33 (1 pl.). GOMES S.R. 2007. Filogenia morfológica de Veronicellidae, filogenia molecular de Colosi e descrição de uma nova espécie para a familia (Mollusca, Gastropoda, Pulmonata). Doctoral thesis, Universidad Federal do Rio Grande do Sul, Porto Alegre [Brazil]. Pp. 175. https://www.lume.ufrgs.br/bitstream/handle/10183/10940/000595534.pdf GOMES S.R. & TOMÉ J.W. 2004. Diversity and distribution of the Veronicellidae (Gastropoda: Soleolifera) in the Oriental and Australian biogeographical regions. Memoirs of the Queensland Museum, 49 (2): 589-601. GREEFF R. 1882. Über die Landschneckenfauna der Insel São Thomé. Zoologische Anzeiger, 5: 516-521. GRIFFITHS O.L. & FLORENS V.F.B. 2006. A field guide to the non-marine molluscs of the Mascarene Islands (Mauritius, Rodrigues and Réunion) and the northern dependencies of Mauritius. Bioculture Press, Maurius. Pp. i-xv, 1-185. GUILLER A. & MADEC L. 2010. Historical biogeography of the land snail Cornu aspersum: a new scenario inferred from haplotype distribution in the Western Mediterranean basin. BMC Evolutionary Biology 2010, 10:18. https://doi.org/10.1186/1471-2148-10-18 HAAS F 1937 Neue und kritische Pupilliden. Archiv für Molluskenkunde, 69: 2-12, pls 1-3.

54

HAUSDORF B. 2007. Revision of the American Pupisoma species (Gastropoda: Pupilloidea). Journal of Natural History, 41 (21-24): 1481-1511. https://doi.org/10.1080/00222930701400169 HAUSDORF B. & BERMÚDEZ C.I.M. 2003. Luntia insignis Smith, 1898, is a synonym of Streptostele (Tomostele) musaecola (Morelet, 1860) (Gastropoda: Streptaxidae) - an African tramp and its distribution in America. Malacologia, 45 (1): 185-187. HERBERT D. & KILBURN D. 2004. Field guide to the land snails and slugs of eastern South Africa. Natal Museum, Pietermaritzburg, South Africa. Pp. 336. HEYNEMANN F.D. 1868. Die Nacktschnecken von der Prinzeninsel. Malakozoologische Blätter, 15: 32-39, pl. 1. HOLYOAK D.T. & HOLYOAK G.A. 2016. Revision of Rhysotina (Gastropoda: Urocyclidae) land-snails endemic to the West African island of São Tomé. Journal of Conchology, 42 (3): 59-72. JONES P. & TYE A. 2006. The birds of São Tomé & Príncipe, with Annobón islands of the Gulf of Guinea. An annotated checklist. British Ornithologists' Union & British Ornithologists' Club, Oxford. Pp. xix + 172. KENNARD A.S. 1942. The Histoire and Prodrome of Férussac. Part II. Proceedings of the Malacological Society of London, 25(3): 105-110 KERNEY M.P. & CAMERON R.A.D. 1979. Land snails of Britain & north-west Europe. HarperCollins, London. 288 pp., 24 pls. LEVENTIS A.P. & OLMOS F. 2009. The birds of São Tomé e Príncipe: a photoguide. Aves & Fotos Editora, São Paulo (Brazil). Pp. 142. MADEC L., BELLIDO A. & GUILLER A. 2003. Shell shape of the land snail Cornu aspersum in North Africa: unexpected evidence of a phylogeographical splitting. Heredity, 91: 224-231. MARTENS E. VON 1860. Die Heliceen nach natürlicher Verwandtschaft systematisch geordnet von Joh. Christ. Albers. Zweite Ausgabe. Engelman, Leipzig. Pp. i-xviii, 1- 359. [original manuscript by Albers edited and published by Martens]. MARTENS E. VON 1876. Die von Prof. Dr. R. Buchholz in Westafrika gesammelten Land- und Süsswasser-Mollusken. Monatsberichte der Königlichen Preussische Akademie des Wissenschaften zu Berlin, 1876: 253-274, pls 1-5. MORELET A. 1848. Testacea quædam Africæ occidentalis terrestria et fluviatilia. Revue Zoologique, par La Société Cuvierienne, 11: 351-355. MORELET A. 1858. Séries conchyliologiques comprenant l'énumération de mollusques terrestres et fluviatiles recueillis pendant le cours de différents voyages, ainsi que la description de plusieurs espèces nouvelles. Klincksieck, Paris. Livr. 1, pp. 1-34, pls 1-3. MORELET A. 1860. Descriptions de nouvelles espèces de l'Afrique occidentale, rapportées par M. le capitaine Vignon. Journal de Conchyliologie, 8 (2): 189-191. MORELET A. 1868. Voyage du Dr Friederich Welwitsch ... dans les Royaumes d'Angola et de Benguella ... Mollusques terrestres et fluviatiles. J.-B. Baillière et fils, Paris. Pp. 102, pls 9 + map. MORELET A. 1873. Novitates conchyliologicae in itinere per Africam aequinoctialiem a Cl. Marche et de Compiègne collectae. Journal de Conchyliologie, 21 (4): 329-332. MÜLLER O.F. 1774. Vermivm terrestrium et fluviatilium, seu animalium, infusoriorum, helminthicorum, et testaceorum, non marinorum, succincta historia. Volumen alterum. Heineck & Faber, Havniæ & Lipsiæ. Pp. xxxvi, 214 [+10]. NOBRE A. 1886. Conchas terrestres e marinhas recolhidas pelo Sr. Adolpho Moller na Ilha de S. Thomé em 1885. Boletim da Sociedade Geographia de Lisboa, (6) 4: 1-15. (Offprint has same page numbers).

55

NOBRE A. 1891. Contribuições para a fauna malacologica da Ilha de S. Thomé (Continuado). O Instituto, Coimbra, (2) 38: 928-935. [Molluscos terrestres on pp. 932-935; offprints with separate pagination]. NOBRE A. 1894. Sur la faune malacologique des îles de S. Thomé et de Madère. Ile de S. Thomé. Annaes de Sciencias naturaes, Publicados por Augusto Nobre, Porto, 1: 91-94, pl. 5. PANISI M. 2017. Biological invasion and the conservation of endemic island species: São Tomé Archachatina giant snails (Pulmonata: Achatinidae). Masters thesis, Universidade de Lisboa, Lisboa [Portugal]. Pp. 75. https://repositorio.ul.pt/handle/10451/31613 PFEIFFER L. 1868. Monographia heliceorum viventium. Sistens descriptiones systematicas et criticas omnium huius familiae generum et specierum hodie cognitarum. Brockhaus, Lipisiae. pp. [1-2], 1-598. PILSBRY H.A. 1902. New land shells of the Japanese Empire. Nautilus, 16: 21-22. PILSBRY H.A. 1904-1905. Manual of Conchology ..., Second Series: Pulmonata. Vol. XVII African Achatinidæ. Conchological Department, Academy of Natural Sciences of Philadelphia, Philadelphia. Pp. xviii + 232, pls. 65. (Part 65, pp. 1-64, pls 1-10, Sept. 23 1904; Part 66, pp. 65-112, pls 11-23, Nov. 26 1904; Part 67, pp. 113-208, pls 24-43, Mar 1 1905; Part 68, pp. 209-232, pls 44-65, June 1905; Title-page, contents & Introduction, June 1905). PILSBRY H.A. 1919. A review of the land mollusks of the Belgian Congo chiefly based on the American Museum Congo Expedition, 1909-1915. Bulletin of the American Museum of Natural History, 40: 1-370. PILSBRY H.A. 1920-1921. Manual of Conchology. Second series: Pulmonata. vol. 26. Pupillidæ (Vertigininæ, Pupillininæ). Conchological Department, Academy of Natural Sciences of Philadelphia, Philadelphia. Pp. iv + 254, pls 24. PILSBRY H.A. 1922-1926. Manual of Conchology. Second series: Pulmonata. vol. 27. Pupillidæ (Orculinæ, Pagodulinæ, Acanthinulinæ, Etc.). Conchological Department, Academy of Natural Sciences of Philadelphia, Philadelphia. Pp. v + 369, pls 32. PRESTON H.B. 1912. Diagnoses of new species of terrestrial and fluviatile shells from British and German East Africa, with the description of a new genus (Eussoia) from the Eusso River, B.E. Africa. Proceedings of the zoological Society of London, 1912: 183-193. RAHEEM D. & NAGGS F. 2006. The Sri Lankan endemic semi-slug Ratnadvipia (: Ariophantidae) and a new species from southwestern Sri Lanka. Systematics and Biodiersity, 4(1): 99-126. RANG S. 1831. Description des coquilles terrestres recueillies pendant un voyage à la côte occidentale d'Afrique, et au Brèsil. Annales des Sciences Naturelles, 24: 5-63, pls 1, 3. REINHARDT 1886. Herr Reinhardt legte eine von Herrn Dr. Gottsche gesammelte neue japanische Helix vor. Sitzungsberichte der Gesellschaft Naturforschender Freunde zu Berlin, for 1886: 115-116. ROWSON B. & HERBERT D. 2016. The type species and circumscription of the species-rich Afrotropical snail genus Gulella L. Pfeiffer 1856, based on anatomical and mtDNA data (Mollusca: : Streptaxidae). Archiv für Molluskenkunde, 245 (1): 69-84. SCHILEYKO A.A. 1999. Treatise on Recent terrestrial pulmonate molluscs. Part 4. Draparnaudiidae, Caryodidae, Macrocyclidae, Acavidae, Clavatoridae, Dorcasiidae, Sculptariidae, Corillidae, Plectopylidae, Megalobulimidae, Strophocheilidae, Cerionidae, Achatinidae, Subulinidae, Glessulidae, Micractaeonidae, Ferrussaciidae. Ruthenica, Supplement, 2: 437-564. SCHILEYKO A.A. 2000. Treatise on Recent terrestrial pulmonate molluscs. Part 6. Rhytididae, Chlamydephoridae, Systrophiidae, Haplotrematidae, Streptaxidae, Spiraxidae, Oleacinidae, Testacellidae. Ruthenica, Supplement, 2: 731-880.

56

SCHILEYKO A.A. 2001. Treatise on Recent terrestrial pulmonate molluscs. Part 7. Endodontidae, Thyrophorellidae, Charopidae. Ruthenica, Supplement, 2: 881-1034. SCHILEYKO A.A. 2002a. Treatise on Recent terrestrial pulmonate molluscs. Part 8. Punctidae, Helicodiscidae, Discidae, Cystopeltidae, Euconulidae, . Ruthenica, Supplement, 2: 1035-1166. SCHILEYKO A.A. 2002b. Treatise on Recent terrestrial pulmonate molluscs. Part 9. Helicarionidae, Gymnarionidae, Rhysotinidae, Ariophantidae. Ruthenica, Supplement, 2: 1167-1307. SMITH E.A. 1890. On a new genus and some new species of shells from Lake Tanganyika. The Annals and Magazine of natural History, (6th ser.) 6: 93-96. SOLEM A. 1966. Some non-marine mollusks from , with notes on the classification of the Helicarionidae. Spolia Zoologica Musei Hauniensis København, 24: 7-110, pls 1-3. SWAINSON W. 1821. Exotic conchology : or, figures and Descriptions of rare, beautiful, or undescribed shells, drawn on stone from the most select specimens ... Vol. 1 (parts 1, 2), William Wood, London. Plates not numbered. TAN S.K. & CLEMENTS R.G. 2011. Limicolaria flammea (Müller, 1774), another potentially invasive African land snail in tropical Asia. Tropical Conservation Science, 5(1): 97-102. TAN S.K. & LOW M.E.Y. 2011. An update on the West African Limicolaria flammea (Müller, 1774) in Singapore, and its distinction from the confamilial Achatina fulica Bowdich, 1822 (Mollusca: Gastropoda: Achatinidae). Nature in Singapore, 4: 311-317. TIELECKE H. 1940. Anatomie, Phylogenie und Tiergeographie der Cyclophoriden. Archiv für Naturgeschichte, (B) N.F. 9: 317-371. TOMLIN J.R. LE B. & SHACKLEFORD L.J. 1914. The marine Mollusca of São Tomé, I. Journal of Conchology, 14(8): 239-256, pl. 4. VAN GOETHEM J. 1977. Révision systématique des Urocyclinae (Mollusca, Pulmonata, Urocyclidae). Annales du Musée Royal de l'Afrique Centrales, Sciences Zoologiques, 218: 1-355, pls 1-4. VAN MOL J.J. 1977. Révision des Urocyclidae (Mollusca, Gastropoda, Pulmonata) Anatomie, systématique, zoogéographie première partie. Annales du Musée Royal de l'Afrique Centrales, Sciences Zoologiques, 180: 1-234. VAN MOL J.J. & BRUGGEN A.C. VAN 1971. Description d'un nouveau genre d'Euconulidae d'Afrique et discussion sur les affinités systématique de cette famille (Mollusca, Gastropoda Pulmonata). Revue de zoologie et de botanique africaines, 84 (3-4): 285- 296. VERDCOURT B. 1961. A new species of Helicarionidae from Portuguese East Africa together with a discussion of the affinities of several other East African species. Journal de Conchyliologie, 101: 120-134. VERDCOURT B. 1991. Some notes on East African endodontoid snails ‒ Part 2. Conchologists' Newsletter, 117: 371-378. VERDCOURT B. 2006. A revised list of the non-marine Mollusca of East Africa (Kenya, Uganda and Tanzania, excluding Malawi). B. Verdcourt, Maidenhead [U.K.]. Pp. 75. VERMEULEN J.J. & WHITTEN A.J. 1998. Fauna Malesiana guide to the land snails of Bali. Backhuys Publishers/Fauna Malesiana Foundation, Leiden. Pp. ix + 164. VIGNON 1888. Catalogue des mollusques terrestres et fluviatiles récoltés sur la côte occidentale d'Afrique [aves des remarques sur ces espèces, par M. C.F. ANCEY]. Bulletins de la Société Malacologique de France, 5: 65-76. WIKTOR A. 2000. (Gastropoda: Pulmonata) – a systematic monograph. Annales Zoologici, Warszawa, 49 (4): 347-590.

57

WIKTOR A. 2001. The slugs of Greece (Arionidae, , , Agriolimacidae - Gastropoda, ). Fauna Graeciae, Natural History Museum of Crete & Hellenic Zoological Society, Irakleio, 8: i-viii, 1-241. WINTER, A.J., DE 1990. On some small west African land snails (Gastropoda: Pulmonata). Journal of Conchology, 33 (5): 305-309. WINTER, A.J., DE 2008. Redefinition of Thapsia Albers, 1860, and description of three more helicarionid genera from western Africa (Gastropoda, Stylommatophora). Zoologische Mededelingen, Leiden, 82 (41): 441-477. WINTER A.J., DE 2017. "Dwarfs among the pygmies": the genus Punctum in western Africa and the Congo Basin, with descriptions of two new species and notes on other Afrotropical punctoids (Gastropoda: Eupulmonata: Punctidae). Archiv für Molluskenkunde, 146 (2): 203-211. WINTER A.J., DE & VAN BRUGGEN A.C. 1992. Systematic position and distribution of the African land snails Afropunctum seminium (Morelet) and 'Guppya' rumrutiensis (Preston) (Mollusca, Gastropoda Pulmonata: Euconulidae), with the description of two new taxa. Proceedings of Koninklijke Nederlandse Akademie van Wettenshchappen, 95 (4): 515- 533. WINTER A.J., DE & GITTENBERGER E. 1998. The land snail fauna of a square kilometre patch of rainforest in southwestern Cameroon: high species richness, low abundance and seasonal fluctuations. Malacologia, 40: 231-250. WINTER A.J., DE & VASTENHOUT, N. 2013. Revision of the Afrotropical land snail genus Avakubia Pilsbry, 1919, with description of Pseudavakubia gen. n. and eleven new species (Gastropoda: Pulmonata: Streptaxidae). African Invertebrates, 54 (2): 605-663. WRONSKI T., GILBERT K., LONG E., MICHÁ B., QUINN R. & HAUSDORF B. 2014. Species richness and meta-community structure of land snails along an altitudinal gradient on Bioko Island, Equatorial Guinea. Journal of Molluyscan Studies, 80: 161-168. YEUNG N.W. & HAYES K.A. 2018. Biodiversity and extinction of Hawaiian land snails: how many are left now and what we must do to conserve them – a reply to Solem (1990). Integrative and Comparative Biology, 58 (6): 1157-1169. DOI: 10.1093/icb/icyo43

58

Table 1. List of terrestrial Mollusca known from the islands of São Tomé and Príncipe. Key: X confirmed record, x unconfirmed by present authors, * newly recorded in present paper, ? occurrence doubtful, ! confirmed by present authors as still present from living or fresh material collected 2013-2019, a introduced, a? probably introduced, e probably single-island endemic, e? doubtful whether single island endemic, † endemic to São Tomé + Príncipe, m only on coastal beaches, SF Subfamily, numbers in brackets refer to notes at end of table.

São Tomé Príncipe CYCLOPHORIDAE J.E. Gray, 1847or MAIZANIIDAE Tielecke, 1940 Afroditropis molleri (Nobre, 1886) X e! - Chondrocyclus inexspectatus G. Holyoak & D. Holyoak sp. nov. - X e!* Maizania furadana G. Holyoak & D. Holyoak sp. nov. X e!* - Thomeomaizania gascoignei G. Holyoak & D. Holyoak sp. nov. X e!* - Thomeomaizania vandellii (Nobre, 1886) X e! - TRUNCATELLIDAE Gray, 1840 Truncatella clathrus R.T. Lowe, 1832 Xm!(1) - Truncatella princeps Dohrn, 1866 - xm ASSIMINEIDAE H. Adams & A. Adams, 1856 Assiminea sp. xm xm (2) ELLOBIIDAE L. Pfeiffer, 1854 SF Melampodinae Stimson, 1851 (1850) Melampus flavus Gmelin, 1791 (3) xm xm Melampus pusillus Gmelin, 1791 xm xm Melampus sp. (4) Xm!* - SF Pedipedinae P. Fischer & Crosse, 1880 Pedipes afer (Gmelin, 1791) xm(5) Pedipes sp. - xm(6) Rafinesque, 1815 Onchidella sp. xm(7) - VERONICELLIDAE J.E. Gray, 1840 Laevicaulis alte (A. Férussac, 1821) Xa!* Xa!* Pseudoveronicella forcarti D. Holyoak, G. Holyoak & F. Sinclair sp. nov. - Xe!* Pseudoveronicella liberiana (Gould, 1850) X! X! Pseudoveronicella thomensis (Girard, 1893) Xe?! - ACHATINIDAE Swainson, 1840 SF Achatininae Archachatina bicarinata (Bruguière, 1792) Xe†! Xe†! Archachatina marginata (Swainson, 1821) Xa! X a! Atopocochlis exaratus (O.F. Müller, 1774) Xe! -(8) Columna columna (O.F. Müller, 1774) - Xe! Columna hainesi L. Pfeiffer, 1856 - xe Columna leai Tryon, 1866 - Xe Lignus alabaster (Rang, 1831) - Xe! Limicolaria flammea (O.F. Müller, 1774) Xa!* - SF Subulininae P. Fischer & Crosse, 1877 Pseudoglessula fuscidula (Morelet, 1858) (9) x x Striosubulina striatella (Rang, 1831) (10) X! X! Subulina feai Germain, 1912 - xe Subulina moreleti Girard, 1893 - Xe! Subulina newtoni Girard, 1893 - Xe! SF Cecilioidinae Mörch, 1864 Cecilioides sp. x?(11) - SF Petriolinae Schileyko, 1999 Bocageia lotophaga (Morelet, 1848) - xe Petriola clava (L. Pfeiffer, 1864) X e! - Petriola marmorea (Reeve, 1850) X e! - Petriola monticola (Morelet, 1866) X e! - SF Pyrgininae Germain, 1916 Pyrgina umbilicata Greeff, 1882 X e! - Thomea newtoni Girard, 1893 X e! - SF Opeatinae Thiele, 1931

59

Opeas dohrni (Girard, 1893) x x Opeas greeffi (Girard, 1893) x X! Opeas pauper (Dohrn, 1866) x† x† Opeas pumilum (Pfeiffer, 1840) X! - Opeas subpauper Germain, 1912 - xe Pseudopeas crossei (Girard, 1893) X!* X! SF Thyrophorellinae Girard, 1895 Thyrophorella thomensis Greeff, 1882 X e! - SF undet. Nothapalus solitarius G. Holyoak & D. Holyoak sp. nov. X e!* - MICRACTAEONIDAE Schileyko, 1999 Micractaeon koptawelilense (Germain, 1934) X!* - STREPTAXIDAE Gray, 1860 SF Enneinae Bourguignat, 1883 Gulella azeitonae D. Holyoak, G. Holyoak & F. Sinclair sp. nov. - Xe!* Gulella crystallum (Morelet, 1848) (12) - Xe! Gulella joubini (Germain, 1912) -(8) xe Gulella sorghum (Morelet, 1848) - Xe! Streptostele abbreviata D. Holyoak, G. Holyoak & F. Sinclair sp. nov. - Xe!* Streptostele fastigiata (Morelet, 1848) -(8) Xe! Streptostele (?) feai Germain, 1912 - xe Streptostele folini (Morelet, 1858) -(13) Xe! Streptostele (?) moreletiana Dohrn, 1866) x† x† Tomostele musaecola (Morelet, 1860) Xa!* - PUNCTIDAE Morse, 1864 Punctum camerunense de Winter, 2017 X!* - CHAROPIDAE Hutton, 1884 SF Charopinae Trachycystis iredalei Preston, 1912 X!* - SUCCINEIDAE Beck, 1837 Quickia concisa (Morelet, 1848) X! X! CERASTIDAE Wenz, 1923 (14) burnayi (Dohrn, 1866) X! X! Rachis dohrni (Greeff, 1882) Xe! - Rachis eminulus (Morelet, 1848) X! X! Rachis hispidus (Greeff, 1882) Xe! - GASTROCOPTIDAE Pilsbry, 1918 Gastrocopta nobrei (Girard, 1893) Xe! - TRUNCATELLINIDAE Steenberg, 1925 Truncatellina thomensis D. Holyoak & G. Holyoak sp. nov. Xe!* - VALLONIIDAE Morse, 1864 (15) Pupisoma dioscoricola (C.B. Adams, 1845) X!* - Pupisoma harpula (Reinhardt, 1886) X!* - AGRIOLIMACIDAE H.Wagner, 1935 Deroceras laeve (O.F. Müller, 1774) Xa!* - EUCONULIDAE H.B. Baker, 1928 Afroconulus roseus D. Holyoak & G. Holyoak sp. nov. Xe!* - Afropunctum seminium (Morelet, 1873) X!* X!* HELICARIONIDAE Bourguignat, 1877 Apothapsia moreleti (Germain, 1915) Xe! - Apothapsia thomensis (Dohrn, 1866) Xe! - UROCYCLIDAE Simroth, 1889 SF Urocyclinae Dendrolimax greeffi Simroth, 1889 Xe! - Dendrolimax heynemanni Heynemann, 1868 - Xe! SF Rhysotininae Schileyko, 2002 Rhysotina hepatizon (Gould, 1845) Xe! - Rhysotina sublaevis G. Holyoak & D. Holyoak, 2016 Xe! - Rhysotina welwitschi (Morelet, 1866) Xe! - SF Sheldoniinae Connolly, 1925 Africarion dumeticola (Dohrn, 1866)(16) - xe? Principicochlea tenuitesta D. Holyoak, G. Holyoak & F. Sinclair sp. nov. - Xe!* Principitrochoidea aglypta (Dohrn, 1866) -(8) Xe!

60

Principitrochoidea convexa G. Holyoak, D. Holyoak & F. Sinclair sp. nov. - Xe!* Principitrochoidea folini (Morelet, 1848) - X! Thomithapsia bomsucessica G. Holyoak & D. Holyoak sp. nov. Xe!* - Thomitrochoidea trindadensis D. Holyoak & G. Holyoak sp. nov. Xe!* - Rafinesque, 1815 SF Helicinae Cornu aspersum (O.F. Müller, 1774) Xa!* -

Notes (1) TOMLIN & SHACKLEFORD (1914: 253), FERNANDES & ROLÁN (1993: 34), CGAH; (2) Listed as new record from both islands by FERNANDES & ROLÁN (1993: 34); (3) CGAH specimens collected in 2018 have been identified as Melampus liberianus H. Adams & A. Adams, 1854; (4) CGAH specimens collected in 2018 have been identified tentatively as Melampus cf. lividus (Deshayes, 1830); (5) TOMLIN & SHACKLEFORD (1914: 240), FERNANDES & ROLÁN (1993: 41); CGAH specimens collected in 2018 have been identified as Pedipes dohrnii d'Ailly, 1896; (6) DOHRN (1866: 133); (7) FERNANDES & ROLÁN (1993: 41); (8) COSSIGNANI (2014: 194-195) figured shells as Ennea joubini, Trochonanina "aglypta" and Streptostele fastigiata with locality as São Tomé and Atopochoclis [sic ]exarata variously as "São Tomé, Guinea Equatoriale" and "São Tomé e Principe", all of which apparently involve errors in the localities; his T. aglypta is a misidentified T. folini; (9) Placed in Ischnoglessula by BREURE ET AL. (2018: 290-291); (10) Stenogyra (Subulina) angustior Dohrn, 1866 is regarded as a synonym of S. striatella because shape and size of shells in some samples collected in 2018 (CGAH) intergrade; (11) Cecilioides sp. was "recently recorded on São Tomé for the first time" (GASCOIGNE 1994b: 798); (12) COSSIGNANI (2014: 194) figured three very similar shells as "Ennea joubini (Germain, 1923), São Tomé, 7 mm". These appear closely similar to Gulella crystallum (Morelet, 1848), endemic on Príncipe. (13) S. folini was listed by GERMAIN (1916: 161) for São Tomé in a table, but perhaps in error for S. Moreleti [sic] which was listed on the following line; (14) Generic allocation of all Cerastidae here is provisional; Buliminus (Rachis) Castroi Nobre, 1891 (NOBRE, 1891: 933; p. 25 in offprint) and Buliminus (Rachis) Crossei Nobre, 1891 (ibid., p. 934) from São Tomé appear to be synonyms of Rachis dohrni and R. eminulus respectively; neurica (Reeve) was listed by GASCOIGNE (1994a: 2) for Príncipe, São Tomé and Bioko, but the basis for this appears to be unclear; (15) A single tiny shell apex from Príncipe collected in 2018 (CGAH) apparently represents a Pupisoma sp., a genus otherwise unrecorded on the island; (16) Based on Vitrina dumeticola Dohrn, 1866 (DOHRN, 1866: 119, pl. 5, figs 1-4) with TL "im Norden der Insel [Príncipe] auf der Pflanzung Azeitona"; the taxon is known only from shells, so its treatment as Helicarion (Africarion) dumeticola Dohrn by GERMAIN (1916: 161) should be regarded as provisional.

61

Figure headings

Figure 1. Shells of terrestrial Mollusca from São Tomé and Príncipe, (1) Cyclophoridae and Maizaniidae. A: Chondrocyclus inexspectatus sp. nov., holotype, Príncipe; B: Thomeomaizania gascoignei sp. nov., holotype, São Tomé; C: Thomeomaizania vandellii, São Tomé, path from above Bom Sucesso Botanical Garden towards Lagoa Amélia, ca N0.2823º E6.5964º, montane forest, 1290-1415 m alt., 9 Dec. 2018, GAH & DTH 18-21, in CGAH; D: Maizania furadana sp. nov., holotype, São Tomé. See text for details of other localities, etc. Figura 1. Conchas de moluscos terrestres de São Tomé e Príncipe, (1) Cyclophoridae and Maizaniidae. A: Chondrocyclus inexspectatus sp. nov., holótipo, Príncipe; B: Thomeomaizania gascoignei sp. nov., holótipo, São Tomé; C: Thomeomaizania vandellii, São Tomé, caminho entre o Jardim Botânico do Bom Sucesso e a Lagoa Amélia, ca N0.2823º E6.5964º, floresta de montanha, 1290-1415 m alt., 9 Dez. 2018, GAH & DTH 18-21, in CGAH; D: Maizania furadana sp. nov., holótipo, São Tomé. Ver texto para detalhes de outras localidades, etc.

Figure 2. Shells of terrestrial Mollusca from São Tomé and Príncipe, (2). A: Nothapalus solitarius sp. nov., holotype, São Tomé; B: Gulella azeitonae sp. nov., holotype, Príncipe; C: Afroconulus roseus sp. nov., holotype, São Tomé; D: ditto, showing umbilical area of base of shell at larger scale; E: Truncatellina thomensis sp. nov., holotype, São Tomé; F: ditto, showing shell aperture at large scale. See text for details of all localities, etc. Figura 2. Conchas de moluscos terrestres de São Tomé e Príncipe, (2). A: Nothapalus solitarius sp. nov., holótipo, São Tomé; B: Gulella azeitonae sp. nov., holótipo, Príncipe; C: Afroconulus roseus sp. nov., holótipo, São Tomé; D: ditto, evidenciando uma ampliação da zona umbilical da base da concha; E: Truncatellina thomensis sp. nov., holótipo, São Tomé; F: ditto, evidenciando uma ampliação da abertura da concha. Ver texto para detalhes de outras localidades, etc.

Figure 3. Shells of terrestrial Mollusca from São Tomé and Príncipe (3). A: Streptostele abbreviata sp. nov., holotype, Príncipe; B: Principicochla tenuitesta sp. nov., holotype, Príncipe; C: Principitrochoidea convexa sp. nov., holotype, Príncipe; D: Thomithapsia bomsucessica sp. nov., holotype, São Tomé; E: Thomitrochoidea trindadensis sp. nov., holotype, São Tomé. See text for details of all localities, etc. Figura 3. Conchas de moluscos terrestres de São Tomé e Príncipe, (3). A: Streptostele abbreviata sp. nov., holótipo, Príncipe; B: Principicochla tenuitesta sp. nov., holótipo, Príncipe; C: Principitrochoidea convexa sp. nov., holótipo, Príncipe; D: Thomithapsia bomsucessica sp. nov., holótipo, São Tomé; E: Thomitrochoidea trindadensis sp. nov., holótipo, São Tomé. Ver texto para detalhes de outras localidades, etc.

Figure 4. Penial anatomy of Pseudoveronicella. A: P. liberiana (São Tomé, by EN2 just NE of Monte Mário "roça", N0.0606° E6.5561°, ca 236 m alt., forest on slopes by road, 13 Dec. 2013, GAH & DTH ST13); B: P. thomensis (São Tomé, path to Lagoa Amélia from Bom Sucesso Botanical Garden, N0.2786° E6.6000°, ca 1300 m alt., in montane forest, 4 Dec. 2013, GAH & DTH ST4) and C and D: P. forcarti sp. nov. (Príncipe, holotype). A and B: show convex dorsal surface of verge; C: shows apex of verge viewed from beyond its tip, dorsal surface uppermost; D: shows view from side, dorsal surface uppermost. Figura 4. Anatomia peniana de Pseudoveronicella. A: P. liberiana (São Tomé, na EN2 a NE de Monte Mário "roça", N0.0606° E6.5561°, ca 236 m alt., floresta no declive junto à estrada, 13 Dez. 2013, GAH & DTH ST13); B: P. thomensis (São Tomé, caminho entre a

62

Lagoa Amélia e o Jardim Botânico do Bom Sucesso, N0.2786° E6.6000°, ca 1300 m alt., em floresta de montanha, 4 Dez. 2013, GAH & DTH ST4) e C e D: P. forcarti sp. nov. (Príncipe, holótipo). A e B: ilustram a superfície dorsal convexa do vértice; C: ilustra o ápice do vértice visto da extremidade da superfície dorsal da margem; D: ilustra vista lateral da extremidade da superfície dorsal.

Figure 5. Genital anatomy of Petriola from São Tomé. A-C: P. clava; A: distal genitalia (just S. of Milagrosa, on road between Trindade and Bombaim, N0.27667° E6.65986°, 398 m alt., tall secondary forest & abandoned cocoa plantations, 2 Dec. 2018, GAH & DTH 18-10); B: longitudinal section of penis to show wall of lumen (path to Lagoa Amélia from Bom Sucesso Botanical Garden, N0.2825° E6.5969°, ca 1400 m alt., montane forest, 7 Dec. 2013, GAH & DTH ST8); C: partial dissection of foreparts of snail with epiphallus extended out of genital pore (path between Bom Sucesso and Macambrará, near Rio Manuel Jorge, N0.2847° E6.6097°, ca 1149 m alt., cultivation edges & small forest patches, 3 Dec. 2013, GAH & DTH ST3); D-F, I: P. marmorea; D: genitalia with attached right ommatophore and associated muscles (Morro Claudina, N0.2912º E6.6057º, 1289 m, montane forest, 24 Nov. 2018, DTH & GAH 18-2); E: detail of penis, vagina and vas deferens (18-2); I: details of distal genitalia and muscle attaching it to right ommatophore (ST8); G, H: P. monticola; G: distal genitalia with attached right ommatophore and associated muscles (Morro Claudina, N0.2918º E6.6059º, 1254-1292 m, montane forest, 27 Nov. 2018, DTH & GAH 18-6); H: details of distal genitalia (by EN2 just NE of Monte Mário "roça", N0.0606° E6.5561°, ca 236 m alt., forest on slopes by road, 13 Dec. 2013, GAH & DTH ST13). Figura 5. Anatomia peniana de Petriola de São Tomé. A-C: P. clava; A: genitália distal (mesmo a S. de Milagrosa, na Estrada entre Trindade e Bombaim, N0.27667° E6.65986°, 398 m alt., floresta secundária alta & plantações de cacau abandonadas, 2 Dez. 2018, GAH & DTH 18-10); B: secção longitudinal do pénis para mostrar a parede do lúmen (caminho entre a Lagoa Amélia e o Jardim Botânico do Bom Sucesso, N0.2825° E6.5969°, ca 1400 m alt., floresta de montanha, 7 Dez. 2013, GAH & DTH ST8); C: disseção parcial da zona frontal do caracol com o epifalo a surgir do poro genital (caminho entre o Bom Sucesso e Macambrará, perto do Rio Manuel Jorge, N0.2847° E6.6097°, ca 1149 m alt., bermas cultivadas & pequenas manchas florestais, 3 Dez. 2013, GAH & DTH ST3); D-F, I: P. marmorea; D: genitália com omatóforo direito anexado e músculos associados (Morro Claudina, N0.2912º E6.6057º, 1289 m, floresta de montanha, 24 Nov. 2018, DTH & GAH 18-2); E: detalhe do pénis, vagina e vaso deferente (18-2); I: detalhes da genitália distal e músculos que o unem ao omatóforo direito (ST8); G, H: P. monticola; G: genitália distal com omatóforo direito anexado e músculos associados (Morro Claudina, N0.2918º E6.6059º, 1254-1292 m, floresta de montanha, 27 Nov. 2018, DTH & GAH 18-6); H: detalhes da genitália distal (na EN2 mesmo a NE de Monte Mário "roça", N0.0606° E6.5561°, ca 236 m alt., floresta no declive junto à estrada, 13 Dez. 2013, GAH & DTH ST13).

Figure 6. Genital anatomy of Streptostele from Príncipe. A-C: S. abbreviata sp. nov. (holotype); A: distal genitalia; B: different view of penis; C: details of "cylindrical sac" and associated structures; D: S. fasciata (path to Santa Joaquina, N1.6157° E7.3976°, 216 m alt., tall native forest, 4 Dec. 2018, GAH, DTH & FS 18-12), distal genitalia and two different views of penis of same snail, rotated; E: S. folini (just S. of Santo António on road to Bela Vista, N1.6300º E7.4177º , ca 24 m alt., cultivation & edges of secondary forest, 4 Dec. 2018, GAH 18-14), distal genitalia and different view of proximal end of penis of same snail. Figura 6. Anatomia genital de Streptostele do Príncipe. A-C: S. abbreviata sp. nov. (holótipo); A: genitália distal; B: diferentes vistas do pénis; C: detalhes do "saco cilíndrico"

63

e estruturas associadas; D: S. fasciata (caminho de Santa Joaquina, N1.6157° E7.3976°, 216 m alt., floresta nativa alta, 4 Dez. 2018, GAH, DTH & FS 18-12), genitália distal e duas vistas rotadas do pénis do mesmo caracol; E: S. folini (mesmo a S. de Santo António na estrada para Bela Vista, N1.6300º E7.4177º , ca 24 m alt., cultivos & margens de floresta secundária, 4 Dez. 2018, GAH 18-14), genitália distal e diferentes vistas do ápice proximal do pénis do mesmo caracól.

Figure 7. Genital anatomy and pattern of body coloration of Afroconulus roseus sp. nov. from São Tomé (all based on holotype). A, B: distal genitalia, viewed from opposite sides; C: detail of penis and epiphallus with epiphallus extended; D: dorsal coloration of body preserved in ims and removed from shell (stipple represents grey coloration; black areas shown as black). Figura 7. Anatomia genital e padrão de coloração corporal de Afroconulus roseus sp. nov. de São Tomé (todos com base no holótipo). A, B: genitália distal, vista de lados opostos; C: detalhe do pénis e epifalo, como o epifalo extendido; D: coloração dorsal do corpo preservado em ims e removido da concha (ponteado representa coloração cinzenta; áreas negras representadas a negro).

Figure 8. Genital anatomy and external structures of bodies of Principicochlea tenuis sp. nov. from Príncipe and Tomithapsia bomsucessica sp. nov. from São Tomé. A-F: P. tenuis (holotype, except E); A: anatomy of distal genitalia; B: right-hand side of preserved specimen (shell removed) to show form of body and mantle laps; C: detail of forepart of right-hand side of body; D: dorsal view of forepart of body of preserved specimen with shell removed and mantle laps extended; E: right-hand side of specimen (paratype) preserved in ims with shell present, the mantle laps retracted into shell; F: partly diagrammatic longitudinal section of atrial appendage. G-J: T. bomsucessica, paratypes; G, H: (site ST4); I, J: (site 18-6); G, I: distal genitalia (the penis retractor muscle broken off in G); H: dorsal view of forepart of body of preserved specimen with shell removed and mantle laps extended (grey pigmentation shown by stippling); J: partly diagrammatic longitudinal section of penis to show pilasters (stippled). Figura 8. Anatomia genital e estruturas externas de corpos de Principicochlea tenuis sp. nov. do Príncipe e Tomithapsia bomsucessica sp. nov. de São Tomé. A-F: P. tenuis (holótipo, exceto E); A: anatomia da genitália distal; B: vista lateral direita de espécime preservado (concha removida) para mostrar a forma do corpo e voltas do manto; C: detalhe da zona frontal direita do corpo; D: vista dorsal da zona frontal do corpo de espécime preservado com a concha removida e as voltas do manto estendidas; E: vista direita de espécime (parátipo) preservado em ims com a concha presente, as voltas do manto retraídas na concha; F: secção longitudinal parcialmente diagramática do apêndice atrial. G-J: T. bomsucessica, parátipos; G, H: (site ST4); I, J: (site 18-6); G, I: genitália distal (músculo retractor do pénis quebrado em G); H: vista dorsal da parte frontal do corpo de espécime com concha removida e voltas do manto estendidas (pigmentação cinzenta ilustrada por ponteado); J: secção longitudinal parcialmente diagramática do pénis para ilustrar as pilastras (ponteado).

Figure 9. Genital anatomy and external features of body of Helicarionoidea. A-C: Apothapsia thomensis (São Tomé, by EN1 well S. of Santa Catarina & inland of NW. coast, N0.2405º E6.4722º, ca 68 m alt., secondary-forest, 1 Dec. 2013, GAH site ST1), all based on same snail; A: distal genitalia; B: detail of penis with coils of epiphallus partly separated and with diagrammatic transverse sections; C: detail of proximal epiphallus only, with coils separated further. D-G: Principitrochoidea convexa sp. nov. (Príncipe, site 18-16A, paratypes); D:

64

lateral view of tail to show caudal horn and diagram of caudal fossa viewed from rear; E: dorsal view of foreparts of preserved body with shell removed and mantle laps spread (also showing white spots on mantle); F: distal genitalia with detail of thickened part of vas deferens; G: partly diagrammatic longitudinal section of penis (showing pilaster on inner wall of lumen with light stipple, muscular thickening of wall of penis with heavier stipple) and two transverse sections. H, I: Principitrochoidea folini (Príncipe, site 18-14, see text); H: distal genitalia; I: detail of penis with outer sheath opened. J, K: Thomitrochoidea trindadensis sp. nov. (holotype, São Tomé, site 18-10); J: distal genitalia; K: diagrammatic longitudinal section of penis to show pilaster (stippled) on inner wall of lumen. Figura 9. Anatomia genital e elementos externos do corpo de Helicarionoidea. A-C: Apothapsia thomensis (São Tomé, na EN1 bem a S. de Santa Catarina & a NW da costa, N0.2405º E6.4722º, ca 68 m alt., floresta secundária, 1 Dez. 2013, GAH site ST1), todos baseados no mesmo caracol; A: genitália distal; B: detalhe do pénis com pregas do epifalo parcialmente separadas e com secções transversais diagramáticas; C: detalhe apenas do epifalo proximal, com as pregas mais separadas. D-G: Principitrochoidea convexa sp. nov. (Príncipe, sítio 18-16A, parátipos); D: vista lateral da cauda para ilustrar a trompa caudal e diagrama da vista posterior da fossa caudal; E: vista dorsal das partes frontais do corpo preservado com concha removida e voltas do manto espalhadas (mostrando igualmente as manchas brancas no manto); F: genitália distal com detalhe de engrossamentos dos vasos deferentes; G: secção longitudinal parcialmente diagramática do pénis (ilustrando pilastra da parede interna do lúmen com ponteado ligeiro, engrossamento muscular da parede peniana com ponteado mais denso) e duas secções transversais. H, I: Principitrochoidea folini (Príncipe, sítio 18-14, ver texto); H: genitália distal; I: detalhe do pénis com baínha externa aberta. J, K: Thomitrochoidea trindadensis sp. nov. (holótipo, São Tomé, sítio 18- 10); J: genitália distal; K: secção longitudinal diagramática do pénis para mostrar pilastra (ponteado) na parede interna do lúmen.

Figure 10. Living terrestrial Mollusca photographed on São Tomé and Príncipe. A, C: Pseudoveronicella thomensis, São Tomé; A: immature, 27 Nov. 2018, Morro Claudina, site 18-6; C: smaller immature, 9 Dec. 2018, by path from Bom Sucesso to Lagoa Amélia (photos © DTH). B, D: Pseudoveronicella liberiana, São Tomé; both 26 Nov. 2018, along track near Ponta Furada, N0.23665° E6.46775° (photos © DTH). E-G: Principicochlea tenuitesta gen. et sp. nov., Príncipe, holotype and paratypes (photos © FS). H-J: Apothapsia thomensis, São Tomé; H: undisturbed snail with mantle laps covering most of shell; I: disturbed subadult snail fleeing, with mantle laps mainly withdrawn into shell; J: threatened snail, body partly withdrawn into shell and flushed with orange-pink colour (photos © RFL). Figura 10. Moluscos terrestres fotografados vivos em São Tomé e Príncipe. A, C: Pseudoveronicella thomensis, São Tomé; A: imaturo, 27 Nov. 2018, Morro Claudina, sítio 18-6; C: imaturo menor, 9 Dez. 2018, no caminho entre o Bom Sucesso e a Lagoa Amélia (fotos © DTH). B, D: Pseudoveronicella liberiana, São Tomé; both 26 Nov. 2018, ao longo do trilho junto a Ponta Furada, N0.23665° E6.46775° (fotos © DTH). E-G: Principicochlea tenuitesta gen. et sp. nov., Príncipe, holótipo e parátipos (fotos © FS). H-J: Apothapsia thomensis, São Tomé; H: caracol relaxado com voltas do manto a cobrir a maior parte da concha; I: caracol subadulto perturbado e em fuga, com voltas do manto maioritariamente retraídas na concha; J: caracol ameaçado, corpo parcialmente retraído na concha e corado de laranja-rosado (fotos © RFL).

65